IEEE Transactions on Cognitive Communications and Networking最新文献_第2页

SCA-LLM: Spectral-Attentive LLM-Based Wireless World Modeling for Agentic Communications 基于频谱关注llm的代理通信无线世界建模

IF 8.6 1区计算机科学

IEEE Transactions on Cognitive Communications and Networking Pub Date : 2026-10-12 DOI: 10.1109/tccn.2026.3741951

Ke He, Le He, Lisheng Fan, Xianfu Lei, Thang X. Vu, George K. Karagiannidis, Symeon Chatzinotas

引用次数: 0

Mobility-Aware Task Offloading and Resource Allocation in SAGIN-based MEC for AIoT: A Deep Reinforcement Learning Approach 基于sagin的AIoT机动感知任务卸载和资源分配：一种深度强化学习方法

IF 8.6 1区计算机科学

IEEE Transactions on Cognitive Communications and Networking Pub Date : 2026-09-29 DOI: 10.1109/tccn.2026.3737898

Wenwu Zhu, Xiaoheng Deng, Wang Li, Shaohua Wan, Geyong Min

引用次数: 0

Cross-Receiver Generalization for RF Fingerprint Identification via Feature Disentanglement and Adversarial Training 基于特征解缠和对抗训练的射频指纹识别跨接收机泛化

IF 8.6 1区计算机科学

IEEE Transactions on Cognitive Communications and Networking Pub Date : 2026-09-28 DOI: 10.1109/tccn.2026.3737891

Yuhao Pan, Xiucheng Wang, Fushuo Huo, Nan Cheng, Wenchao Xu

引用次数: 0

Digital Twin-Enabled Fair and Efficient Resource Allocation for LEO Satellite Networks 基于数字双子星的低轨道卫星网络资源公平高效分配

IF 8.6 1区计算机科学

IEEE Transactions on Cognitive Communications and Networking Pub Date : 2026-08-18 DOI: 10.1109/tccn.2026.3725001

Jinkai Zheng, Tom H. Luan, Guanjie Li, Yao Zhang

引用次数: 0

CoMQT: Complex-Valued Multi-Codebook Discrete Tokenization with VAT for Semi-Supervised Modulation Recognition 半监督调制识别的复值多码本离散令牌化

IF 8.6 1区计算机科学

IEEE Transactions on Cognitive Communications and Networking Pub Date : 2026-07-16 DOI: 10.1109/tccn.2026.3714066

Xunbin Deng, Sheng Wu, Wei Li, Hamada Esmaiel, Mingzhang Zhou, Haixin Sun

引用次数: 0

Energy Efficiency in Quantized MIMO-NOMA Communication Systems 量化MIMO-NOMA通信系统的能量效率

IF 7 1区计算机科学

IEEE Transactions on Cognitive Communications and Networking Pub Date : 2026-03-03 DOI: 10.1109/TCCN.2026.3670130

Thanh Phung Truong;Phuong Nam Tran;Junsuk Oh;Donghyun Lee;Van Dat Tuong;Nhu-Ngoc Dao;Sungrae Cho

{"title":"Energy Efficiency in Quantized MIMO-NOMA Communication Systems","authors":"Thanh Phung Truong;Phuong Nam Tran;Junsuk Oh;Donghyun Lee;Van Dat Tuong;Nhu-Ngoc Dao;Sungrae Cho","doi":"10.1109/TCCN.2026.3670130","DOIUrl":"10.1109/TCCN.2026.3670130","url":null,"abstract":"Reducing power consumption while maintaining high spectral efficiency is crucial in modern communication systems, especially given the rapid expansion of network services and devices. To catch up on this challenge, we explore an energy-efficiency problem in an uplink quantized MIMO (Multiple-Input Multiple-Output) system empowered by NOMA (Non-Orthogonal Multiple Access) in this work. In this context, employing low-resolution quantizers can significantly reduce power consumption, while NOMA enhances spectral efficiency in MIMO systems. Here, we propose a post-actor-added deep deterministic policy gradient (P-DDPG) algorithm that incorporates a novel post-actor process into the DDPG training algorithm to optimize the users’ precoding and base station’s detection matrices while considering their constraints. By leveraging the P-DDPG algorithm, we aim to enhance the system’s energy efficiency, boosting spectral efficiency while minimizing power consumption. Additionally, we design a power-exhaustive searching function added to the trained model in the inference process, ensuring device diversity. Numerical results confirm the algorithm’s convergence and demonstrate its superior performance under various environmental conditions compared to benchmark schemes. Furthermore, we thoroughly analyze the impact of low-resolution quantizers on system performance.","PeriodicalId":13069,"journal":{"name":"IEEE Transactions on Cognitive Communications and Networking","volume":"12 ","pages":"6629-6642"},"PeriodicalIF":7.0,"publicationDate":"2026-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147350420","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Sparse Gaussian–Markov Modeling for Robust and Trustworthy Unknown Cyber Defense 鲁棒可信未知网络防御的稀疏高斯马尔可夫建模

IF 7 1区计算机科学

IEEE Transactions on Cognitive Communications and Networking Pub Date : 2026-02-27 DOI: 10.1109/TCCN.2026.3668824

Chao Zha;Tian Liu;Chungang Lin;Bing Bai;Ruyun Zhang

{"title":"Sparse Gaussian–Markov Modeling for Robust and Trustworthy Unknown Cyber Defense","authors":"Chao Zha;Tian Liu;Chungang Lin;Bing Bai;Ruyun Zhang","doi":"10.1109/TCCN.2026.3668824","DOIUrl":"10.1109/TCCN.2026.3668824","url":null,"abstract":"Network intrusion detection systems (NIDS) are essential for enhancing network security by detecting and responding to attacks promptly. NIDS based on artificial intelligence (AI) offer better adaptability to unknown (out-of-distribution (OOD)) attacks compared to traditional anomaly detection methods. However, many existing AI-based NIDS directly adopt methods from other domains without adequately considering OOD data, which often leads to severe overfitting in trained classifiers and insufficient model reliability and robustness. This issue largely arises because most approaches focus solely on the relationship between individual features and predictions while neglecting the interactions among features, resulting in overfitting and overconfident predictions. Moreover, the retention of certain weak parameters may cause the model to maintain artificially high confidence when confronted with OOD data. To address this, we propose a Gaussian Markov Random Field that better captures traffic data distribution and feature correlations. Furthermore, our GMRF-based classification model uses sparsification with regularization <inline-formula> <tex-math>$L_{0}$ </tex-math></inline-formula> to remove weak parameters. <inline-formula> <tex-math>$L_{0}$ </tex-math></inline-formula> regularization counts the number of non-zero parameters and penalizes model complexity, thereby forcing insignificant weights to become exactly zero and improving generalization. This prevents high prediction probabilities for unknown attacks and effectively distinguishes them from benign traffic. Finally, we evaluated our method in the CICIDS-2017 and CSE-CICIDS-2018 datasets, achieving recall rates of 57% and 77% for unknown attacks while maintaining a recall of more than 95% for known traffic. Our approach significantly outperforms existing SOTA methods, and performance tests on a low-specification machine show a low detection latency of approximately 130 microseconds, making it suitable for real-time applications.","PeriodicalId":13069,"journal":{"name":"IEEE Transactions on Cognitive Communications and Networking","volume":"12 ","pages":"6524-6538"},"PeriodicalIF":7.0,"publicationDate":"2026-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147319586","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

FedFold: Efficient Federated Learning for Inconsistent Dual Resource Heterogeneity FedFold：不一致双资源异构的高效联邦学习

IF 7 1区计算机科学

IEEE Transactions on Cognitive Communications and Networking Pub Date : 2026-02-27 DOI: 10.1109/TCCN.2026.3668835

Tzu-Hsuan Peng;Chieh-Chi Yang;Chih-Hsin Liu;Cheng-Wei Huang;Cheng-Hsien Hsieh;Ai-Chun Pang

{"title":"FedFold: Efficient Federated Learning for Inconsistent Dual Resource Heterogeneity","authors":"Tzu-Hsuan Peng;Chieh-Chi Yang;Chih-Hsin Liu;Cheng-Wei Huang;Cheng-Hsien Hsieh;Ai-Chun Pang","doi":"10.1109/TCCN.2026.3668835","DOIUrl":"10.1109/TCCN.2026.3668835","url":null,"abstract":"With the continuous evolution of 6G networks and edge computing, the demand for intelligent, secure, and efficient distributed learning frameworks is growing. Federated Learning (FL) has become a key approach for privacy-preserving model training, allowing edge devices to collaboratively train a global model without exchanging raw data. However, real-world FL deployments face significant challenges due to system heterogeneity. Devices vary in computational power and network conditions, leading to the straggler effect, where weaker devices slow down training efficiency. While prior works have addressed either computational or communication heterogeneity, they often overlook the interdependent nature of these two factors. Inconsistent capabilities between computation and communication introduce a cross-resource bottleneck: a strong processor with weak bandwidth leads to stalled updates, while a fast network with limited compute yields poor-quality contributions. Focusing on only one resource dimension often pushes the bottleneck to the other, leaving the straggler effect unresolved. To address this issue, we propose Federated Folding (FedFold), a novel FL framework designed to balance dual-resource efficiency in heterogeneous edge environments. FedFold integrates a progressive model splitting strategy to dynamically distribute training loads and local aggregation-based compression to reduce communication overhead. Unlike existing methods that address only one bottleneck at a time, FedFold adapts to inconsistent and dual resource heterogeneity among clients, ensuring robust and scalable learning. The experiments demonstrate that FedFold significantly reduces training time and communication costs while maintaining high model accuracy, outperforming state-of-the-art FL frameworks by over to 70% in efficiency improvements under non-IID scenarios.","PeriodicalId":13069,"journal":{"name":"IEEE Transactions on Cognitive Communications and Networking","volume":"12 ","pages":"6539-6550"},"PeriodicalIF":7.0,"publicationDate":"2026-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147319582","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Personalized Federated Learning With Mixture-of-Experts for Automatic Modulation Classification Under Data Heterogeneity 数据异构下基于专家混合的个性化联邦学习自动调制分类

IF 7 1区计算机科学

IEEE Transactions on Cognitive Communications and Networking Pub Date : 2026-02-27 DOI: 10.1109/TCCN.2026.3668838

Jitong Ma;Jianing Li;Sen Zhao;Hui Dong Li;Jun Yan Dai;Jie Wang;Qiang Cheng

{"title":"Personalized Federated Learning With Mixture-of-Experts for Automatic Modulation Classification Under Data Heterogeneity","authors":"Jitong Ma;Jianing Li;Sen Zhao;Hui Dong Li;Jun Yan Dai;Jie Wang;Qiang Cheng","doi":"10.1109/TCCN.2026.3668838","DOIUrl":"10.1109/TCCN.2026.3668838","url":null,"abstract":"Automatic Modulation Classification (AMC) is a pivotal task for cognitive and intelligent communication systems. Existing AMC methods predominantly rely on centralized deep learning, necessitating the collection and centralized training of modulation signals from various terminals to recognize modulation types. However, in practical applications, there are still two main challenges: 1) signal data are often distributed across different devices and data privacy and security constraints often prohibit centralized data aggregation; 2) Complex and heterogeneous wireless environments lead to Non-Independent and Identically Distributed (Non-IID) data and missing-class phenomena across terminals, rendering a unified global model ineffective for personalized recognition. To address these limitations, this paper proposes a heterogeneous personalized federated learning framework based on a Mixture-of-Experts (MoE) model, termed pFedMixMoE, for the AMC task. Specifically, the proposed framework employs a dynamic gating mechanism to adaptively determine the fusion ratio between a globally shared expert for generalization and client-specific local experts for personalization, thus effectively balancing global knowledge transfer and tailored local adaptation. Furthermore, we design a QKV-based frequency-domain aggregation strategy to improve robustness under heterogeneous environments. Extensive Experimental results demonstrate that pFedMixMoE enhances recognition accuracy by 17.43% and 28.32% compared to mainstream baseline methods under various data distribution and sample type scenarios. These results validate the superior effectiveness and robustness of pFedMixMoE and provide new ideas for constructing personalized models in wireless networks.","PeriodicalId":13069,"journal":{"name":"IEEE Transactions on Cognitive Communications and Networking","volume":"12 ","pages":"6478-6491"},"PeriodicalIF":7.0,"publicationDate":"2026-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147319584","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Energy-Aware and Loss-Resilient Data Collection for AAV-Assisted Unbalanced IoT Networks 无人机辅助不平衡物联网网络的能量感知和损失弹性数据收集

IF 7 1区计算机科学

IEEE Transactions on Cognitive Communications and Networking Pub Date : 2026-02-27 DOI: 10.1109/TCCN.2026.3668820

Yingying Ren;Shaolin Liu;Chaohong Tan;Sijing Duan;Zuqiang Meng;Yongxin Liu;Anfeng Liu;Houbing Herbert Song

{"title":"Energy-Aware and Loss-Resilient Data Collection for AAV-Assisted Unbalanced IoT Networks","authors":"Yingying Ren;Shaolin Liu;Chaohong Tan;Sijing Duan;Zuqiang Meng;Yongxin Liu;Anfeng Liu;Houbing Herbert Song","doi":"10.1109/TCCN.2026.3668820","DOIUrl":"10.1109/TCCN.2026.3668820","url":null,"abstract":"In infrastructure-limited Internet of Things (IoTs) networks, sensing devices are sparsely and irregularly distributed, with limited or intermittent access. Transmiting the data to cloud in real-time is unreliable due to constrained connectivity, harsh environmental conditions, or the energy limitations of nodes. Autonomous Aerial Vehicles (AAVs) offer an energy-efficient solution by periodically collecting sensing data via short-range communication. To address energy constraints and data loss in unbalanced IoTs networks, we propose DQGDC, an energy-aware and loss-resilient AAV data collection framework. DQGDC jointly optimizes hovering point deployment, flight trajectory, and missing data recovery to improve efficiency under energy limitations. Firstly, a clustering and particle swarm optimization-based hovering point design minimizes the number of AAV hover points and redundantly covered nodes. Secondly, a graph convolutional network and double deep Q-Network-based trajectory planner enables adaptive, energy-efficient flight paths. Thirdly, to enhance resilience against sensing loss from inaccessible or unvisited nodes, we develop a Q-learning driven deep nonnegative matrix factorization model for reconstructing missing sensing data. Extensive experiments on realistic and simulated IoTs scenarios demonstrate that DQGDC effectively balances AAV energy consumption and data collection completion.","PeriodicalId":13069,"journal":{"name":"IEEE Transactions on Cognitive Communications and Networking","volume":"12 ","pages":"6551-6564"},"PeriodicalIF":7.0,"publicationDate":"2026-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147319612","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0