Intelligent and Converged Networks最新文献_第3页

Total Contents 全部内容

Intelligent and Converged Networks Pub Date : 2025-12-29 DOI: 10.23919/TUP-ICN.2025.11318126

引用次数: 0

Collaborative DNN Inference in Maritime Edge Intelligence Networks with Group Neural Multi-Armed Bandits 基于群神经多武装强盗的海上边缘情报网络协同DNN推理

Intelligent and Converged Networks Pub Date : 2025-12-29 DOI: 10.23919/ICN.2025.0025

Yulei Wang;Shishi Liu;Di Bai;Yongqiang Cui

{"title":"Collaborative DNN Inference in Maritime Edge Intelligence Networks with Group Neural Multi-Armed Bandits","authors":"Yulei Wang;Shishi Liu;Di Bai;Yongqiang Cui","doi":"10.23919/ICN.2025.0025","DOIUrl":"https://doi.org/10.23919/ICN.2025.0025","url":null,"abstract":"Collaborative Deep Neural Networks (DNNs) inference has emerged as a promising paradigm for growing number of artificial intelligence-integrated maritime Internet of Things (IoT) devices in maritime edge intelligence networks. However, the resource constraints of devices, the delay-sensitive nature of tasks, and the dynamic environmental conditions present significant challenges. While Multi-Armed Bandit (MAB) algorithms have been explored for task offloading, their performance is often constrained in highly dynamic scenarios with complex, nonlinear utility dependencies. To address these challenges, we propose a Group Neural MAB (GN-MAB) approach that jointly optimizes idle device selection (i.e., arm groups) and DNN partitioning decisions (i.e., arms) for efficient collaborative inference. Building upon the neural upper confidence bound algorithm, GN-MAB dynamically balances the exploration and exploitation, enabling continuous adaptation of offloading strategies across sequential inference tasks. Extensive experimental results show that GN-MAB outperforms baseline approaches, achieving superior inference performance while exhibiting robust adaptability to the fluctuating conditions of maritime environments.","PeriodicalId":100681,"journal":{"name":"Intelligent and Converged Networks","volume":"6 4","pages":"378-391"},"PeriodicalIF":0.0,"publicationDate":"2025-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11318124","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145847760","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Fine-Grained Fashion Classification with Limited Labeled Data Via Generative Augmentation 基于生成增强的有限标签数据细粒度时尚分类

Intelligent and Converged Networks Pub Date : 2025-12-29 DOI: 10.23919/ICN.2025.0028

Hongbi Jeong;Zuobin Xiong;Haram Kang;Kyungtae Kang;Junggab Son

{"title":"Fine-Grained Fashion Classification with Limited Labeled Data Via Generative Augmentation","authors":"Hongbi Jeong;Zuobin Xiong;Haram Kang;Kyungtae Kang;Junggab Son","doi":"10.23919/ICN.2025.0028","DOIUrl":"https://doi.org/10.23919/ICN.2025.0028","url":null,"abstract":"Fine-grained classification of fashion items is essential for enhancing user experiences in online shopping, enabling more effective categorization and personalized recommendations. While recent advances in Artificial Intelligence (AI) have shown promise, achieving high classification accuracy typically requires large volumes of labeled data—an expensive and labor-intensive requirement, particularly burdensome for smaller retailers. Existing approaches attempt to mitigate this challenge by pretraining models on synthetic or geometric data and fine-tuning on limited fashion images, but these methods have thus far plateaued at around 90% accuracy, falling short of practical deployment standards. In this paper, we introduce a novel iterative image generation framework designed to overcome data scarcity and significantly improve classification accuracy. Our method combines a conditional Generative Adversarial Network (cGAN) with a ResNet50-based image classifier in a closed-loop system. The cGAN generates synthetic fashion images conditioned on class labels, while the classifier filters outputs based on confidence scores and predefined criteria. This cycle is repeated iteratively, progressively enriching the training dataset with high-quality synthetic images and refining the classifier. We validate our approach on a task involving classification of five distinct neckline types. The converged model achieves an average accuracy of 94.6%, substantially outperforming previous methods despite using a limited amount of real labeled data. These results highlight the effectiveness of iterative data augmentation using generative models for fine-grained visual classification in resource-constrained settings.","PeriodicalId":100681,"journal":{"name":"Intelligent and Converged Networks","volume":"6 4","pages":"311-325"},"PeriodicalIF":0.0,"publicationDate":"2025-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11318125","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145847812","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Dynamic Content Popularity Prediction Based on Spatial-Temporal Feature in Vehicular Networks 基于时空特征的车载网络动态内容流行度预测

Intelligent and Converged Networks Pub Date : 2025-12-29 DOI: 10.23919/ICN.2025.0029

Ying Shi;Yuxia Zang;Tong Wang;Yiping Chen;Chi Feng

{"title":"Dynamic Content Popularity Prediction Based on Spatial-Temporal Feature in Vehicular Networks","authors":"Ying Shi;Yuxia Zang;Tong Wang;Yiping Chen;Chi Feng","doi":"10.23919/ICN.2025.0029","DOIUrl":"https://doi.org/10.23919/ICN.2025.0029","url":null,"abstract":"Content popularity is crucial for edge caching in vehicular networks, where prediction accuracy directly impacts communication performance. To enhance prediction accuracy by fully exploiting the characteristics of content popularity, we propose a novel approach that integrates spatial correlation among roadside units and the temporal correlation of vehicle content requests. We jointly construct a spatial-temporal feature matrix using Euclidean distance and Spearman correlation, called the ES matrix, as the adjacency matrix of the graph structure. This ES matrix is integrated with Spatial-Temporal Transformer Network (STTN) to build the ES-STTNS model. To evaluate its effectiveness, we design a dynamic caching strategy that continuously updates RSU cache in real-time based on predicted content popularity, aiming to maximize Cache Hit Ratio (CHR). Extensive experiments demonstrate that ES-STTNS achieves highly accurate predictions for real vehicular content requests, reducing Mean Absolute Error (MAE) by up to 24.4% and Root Mean Square Error (RMSE) by up to 20.4% compared to temporal-only models, especially over longer prediction horizons, while achieving 90% CHR at higher cache capacities.","PeriodicalId":100681,"journal":{"name":"Intelligent and Converged Networks","volume":"6 4","pages":"326-341"},"PeriodicalIF":0.0,"publicationDate":"2025-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11318129","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145847842","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Real-Time Detection of Railway Signal Gantries via Improved RT-DETR: Edge Deployment and Cloud Empowerment 基于改进RT-DETR的铁路信号门的实时检测：边缘部署和云授权

Intelligent and Converged Networks Pub Date : 2025-12-29 DOI: 10.23919/ICN.2025.0026

Yanbin Weng;Peixin Xie;Xiahu Chen;Hui Xiang;Fukang Chen;Manlu Liu

{"title":"Real-Time Detection of Railway Signal Gantries via Improved RT-DETR: Edge Deployment and Cloud Empowerment","authors":"Yanbin Weng;Peixin Xie;Xiahu Chen;Hui Xiang;Fukang Chen;Manlu Liu","doi":"10.23919/ICN.2025.0026","DOIUrl":"https://doi.org/10.23919/ICN.2025.0026","url":null,"abstract":"The precise and timely extraction of railway signals is crucial for the creation of railway electronic maps. This paper introduces a novel real-time detection approach for dynamically adjusting railway signals, leveraging an enhanced Real-Time DEtection TRansformer (RT-DETR) model. The enhancement involves the integration of a vision Transformer with Dynamically Quantifiable Sampling Attention Mechanism (DQSAM) into the ResNet50 backbone of the RT-DETR framework, thereby enhancing the model's efficiency and accuracy in handling intricate visual tasks. Secondly, an ultra-lightweight and effective Dynamic Grouping upSampler (DyGSample) is inserted into the efficient hybrid encode module as the up-sampling part. This operator can effectively upsample the feature graph without increasing the computational burden, and improve the model resolution and detail capture ability. In addition, in order to solve the problem of deep layer of model network and high operating cost, a new bounding box similarity loss function of rotation intersection over union based on minimum point distance is adopted in this paper, which takes into account all relevant factors of existing loss functions, namely overlapping or non-overlapping regions, center point distance, width and height deviation, and simplifies the calculation process. As a lightweight signal detection model with ultra-fast, high real-time, and high precision, the detection accuracy of this method is improved from 90.21% to 97.45%, which proves the superior performance and effectiveness of the improved real-time dynamic adjustment RT-DETR model in railway signal extraction.","PeriodicalId":100681,"journal":{"name":"Intelligent and Converged Networks","volume":"6 4","pages":"289-310"},"PeriodicalIF":0.0,"publicationDate":"2025-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11318127","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145847814","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Joint Information Freshness Optimization for Digital Twin Communication Infrastructure Through Hierarchical Deep Reinforcement Learning 基于层次深度强化学习的数字孪生通信基础设施联合信息新鲜度优化

Intelligent and Converged Networks Pub Date : 2025-12-29 DOI: 10.23919/ICN.2025.0021

Chenyu Wang;Akshita Maradapu Vera Venkata Sai;Kehua Wang;Yingshu Li

{"title":"Joint Information Freshness Optimization for Digital Twin Communication Infrastructure Through Hierarchical Deep Reinforcement Learning","authors":"Chenyu Wang;Akshita Maradapu Vera Venkata Sai;Kehua Wang;Yingshu Li","doi":"10.23919/ICN.2025.0021","DOIUrl":"https://doi.org/10.23919/ICN.2025.0021","url":null,"abstract":"In recent years, Digital Twin (DT) has emerged as a transformative paradigm for enabling the future of the Internet of Things. By mapping the real-time status of physical entities to their digital counterparts, DTs facilitate the creation of high-fidelity, interactive environments suitable for advanced simulation and deeper insight. One of the key challenges lies in achieving a sufficient level of convergence between Physical Twins (PTs) and their corresponding DTs. To tackle this challenge, we introduce a mobile edge computing environment that enables the coordination between PTs and DTs in Digital Twin Networks (DTNs) by offloading data transmission and processing tasks to the edge. A Hierarchical Deep Reinforcement Learning (HDRL) framework is proposed to improve DT-PT synchronization and enhance coordination effectiveness in optimizing information freshness across multiple action policies within the Digital Twin Communication Infrastructure (DTCI). Our approach is validated through a DTCI simulator, where comprehensive evaluations of age of information performance are conducted. Experimental results show that our HDRL-based solution significantly enhances the information freshness under constrained DTCI resources and across diverse environmental conditions.","PeriodicalId":100681,"journal":{"name":"Intelligent and Converged Networks","volume":"6 4","pages":"265-277"},"PeriodicalIF":0.0,"publicationDate":"2025-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11318128","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145847810","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Two-Stage Sea Level Anomaly Prediction Framework with Dual-Branch Feature Representation 基于双分支特征表示的两阶段海平面异常预测框架

Intelligent and Converged Networks Pub Date : 2025-12-29 DOI: 10.23919/ICN.2025.0023

Kefan Wang;Qi Kang

{"title":"Two-Stage Sea Level Anomaly Prediction Framework with Dual-Branch Feature Representation","authors":"Kefan Wang;Qi Kang","doi":"10.23919/ICN.2025.0023","DOIUrl":"https://doi.org/10.23919/ICN.2025.0023","url":null,"abstract":"Sea Level Anomaly (SLA) is commonly employed to identify mesoscale ocean eddies, and its spatiotemporal prediction is useful for analyzing oceanic dynamic processes. Traditional SLA prediction models often operate directly on raw SLA data, which hinders the extraction of feature representations under complex oceanic dynamics, thereby limiting their predictive accuracy. To overcome this limitation, this paper proposes a two-stage SLA prediction framework with dual-branch feature representation, including a reconstruction stage and a prediction stage. In the first stage, a Dual-Branch Encoder (DB-Encoder) and a Spatially Adaptive Normalization Decoder (S-Decoder) are designed to extract latent feature representations. The DB-Encoder processes raw SLA data and one-hot encodings of iso-valued regions through two separate branches to enhance topological awareness. Then, Sparse regression based Vector Quantization (SVQ) is employed to represent features as weighted combinations of codebook vectors, effectively reducing noise and enhancing robustness. The S-Decoder further refines the reconstruction by applying spatially adaptive normalization, making the output more consistent with realistic ocean dynamics. During the second stage, features are extracted from each frame using the pretrained DB-Encoder and codebooks, and then passed to spatiotemporal prediction networks. The S-Decoder subsequently reconstructs these predicted features into the original spatial representation. Results from the South China Sea show that the proposed method improves both the reconstruction and prediction quality of SLA, while also achieving improved performance in eddy detection tasks, validating its effectiveness and potential for practical applications.","PeriodicalId":100681,"journal":{"name":"Intelligent and Converged Networks","volume":"6 4","pages":"342-357"},"PeriodicalIF":0.0,"publicationDate":"2025-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11318131","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145847538","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Adversarial Strategy for Swarm UAVs Gaming with Real-Time Interaction 具有实时交互的蜂群无人机博弈对抗策略

Intelligent and Converged Networks Pub Date : 2025-09-29 DOI: 10.23919/ICN.2025.0014

Zhe Yang;Jiehong Wu

{"title":"Adversarial Strategy for Swarm UAVs Gaming with Real-Time Interaction","authors":"Zhe Yang;Jiehong Wu","doi":"10.23919/ICN.2025.0014","DOIUrl":"https://doi.org/10.23919/ICN.2025.0014","url":null,"abstract":"The Unmanned Aerial Vehicle (UAV) swarm confrontation problem has become a hot research topic in the field of intelligent control systems in recent years, possessing significant value in artificial intelligence and environmental monitoring and protection. To address the issue of unstable striking ability in battlefield environments, leading to combat inefficiency, we propose a multi-objective optimization game-theoretic confrontation method. This method establishes a dynamic game model based on pure strategy gains with the target. We develop a real-time interactive Laplace matrix to enhance synergy between our UAV swarms. Additionally, a topological method is employed to achieve stable control by designing a real-time mutual interaction synergy structure based on mission state. Finally, the efficient encoding and decoding capabilities of the Transformer model, combined with the multi-intelligence deep deterministic policy gradient algorithm, improve the model's ability to handle state space representations while enabling the UAV swarm to better understand the complex interactions between intelligences. Simulations are conducted using an adversarial decision-making algorithm based on game theory and cooperative communication structures in 2v2 simple scenarios and 4v4 complex scenarios. Compared with the benchmark algorithm, the algorithm in this paper converges faster, is more robust, and takes less time to win the confrontation.","PeriodicalId":100681,"journal":{"name":"Intelligent and Converged Networks","volume":"6 3","pages":"179-194"},"PeriodicalIF":0.0,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11184478","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145183933","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Hardware-Accelerated Real-Time FIR Filtering for Audio Applications Using FPGAs 使用fpga的音频应用硬件加速实时FIR滤波

Intelligent and Converged Networks Pub Date : 2025-09-29 DOI: 10.23919/ICN.2025.0017

Chenmu Ji;Weibei Fan

{"title":"Hardware-Accelerated Real-Time FIR Filtering for Audio Applications Using FPGAs","authors":"Chenmu Ji;Weibei Fan","doi":"10.23919/ICN.2025.0017","DOIUrl":"https://doi.org/10.23919/ICN.2025.0017","url":null,"abstract":"Real-time digital audio processing plays a crucial role in modern electronic systems, such as mobile devices, communication platforms, and multimedia applications. However, ensuring high-performance, low-latency processing under hardware constraints remains a significant challenge, particularly when integrating audio codec chips with FPGA platforms. This paper presents the design and implementation of a real-time digital audio processing system using field-programmable gate array (FPGA) technology and the WM8731 audio codec. Firstly, a robust Inter-Integrated Circuit (I<sup>2</sup>C) interface is developed to configure the WM8731 codec, ensuring accurate initialization and stable operation. Secondly, a serial-to-parallel adaptor (s2p_adaptor) is designed to convert inter-IC sound (I<sup>2</sup>S) serial audio data into parallel format for digital processing, synchronized with bit and frame clocks. Finally, an 8-tap finite impulse response (FIR) filter is implemented using very high speed integrated circuit (VHSIC) hardware description language (VHDL) to enhance audio quality by suppressing high-frequency noise. All modules are synthesized in Quartus II and verified through ModelSim simulations and practical testing on an Altera DE1 development board. Timing diagrams based on the WM8731 datasheet and internal clock signals ensure precise synchronization. Experimental results demonstrate the effectiveness of the system, while final analysis highlights areas for further optimization and future improvements. Beyond audio applications, the proposed architecture is also adaptable to other real-time signal processing tasks, such as biomedical monitoring, radar front-end filtering, and image preprocessing. This extensibility enhances the system's relevance in broader embedded and communication contexts.","PeriodicalId":100681,"journal":{"name":"Intelligent and Converged Networks","volume":"6 3","pages":"209-222"},"PeriodicalIF":0.0,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11184477","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145183972","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Optimum Scheduling of Truck-Based Mobile Energy Couriers (MEC) Using Deep Deterministic Policy Gradient 基于深度确定性策略梯度的汽车移动能源配送优化调度

Intelligent and Converged Networks Pub Date : 2025-09-29 DOI: 10.23919/ICN.2025.0016

Yaze Li;Jingxian Wu;Yanjun Pan

{"title":"Optimum Scheduling of Truck-Based Mobile Energy Couriers (MEC) Using Deep Deterministic Policy Gradient","authors":"Yaze Li;Jingxian Wu;Yanjun Pan","doi":"10.23919/ICN.2025.0016","DOIUrl":"https://doi.org/10.23919/ICN.2025.0016","url":null,"abstract":"We propose a new architecture of truck-based mobile energy couriers (MEC) for power distribution networks with high penetration of renewable energy sources (RES). Each MEC is a truck equipped with high-density inverters, converters, capacitor banks, and energy storage devices. The MEC platform can improve the flexibility, resilience, and RES hosting capability of a distribution grid through spatial-temporal energy reallocation based on the stochastic behaviors of RES and loads. The employment of MEC necessitates the development of complex scheduling and control schemes that can adaptively cope with the dynamic natures of both the power grid and the transportation network. The problem is formulated as a non-convex optimization problem to minimize the total generation cost, subject to the various constraints imposed by conventional and renewable energy sources, energy storage, and transportation networks, etc. The problem is solved by combining optimal power flow (OPF) with deep reinforcement learning (DRL) under the framework of deep deterministic policy gradient (DDPG). Simulation results demonstrate that the proposed MEC platform with DDPG can achieve significant cost reduction compared to conventional systems with static energy storage.","PeriodicalId":100681,"journal":{"name":"Intelligent and Converged Networks","volume":"6 3","pages":"195-208"},"PeriodicalIF":0.0,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11184475","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145183941","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0