Unmanned aerial vehicle (UAV) networks have emerged as promising enablers in sixth generation (6G) communication system because they can support delay-sensitive and energy-constrained applications. However, the limited resources of UAVs and the high computational complexity of traditional methods complicate task offloading and position optimization. At scale, the task offloading and position optimization decisions yield non-stationary interactions among many agents, while standard multi-agent deep reinforcement learning (MADRL) suffers from poor scalability as the joint action space grows exponentially with the number of UAVs. We formulate joint task offloading and 2D position control as a Markov game that minimizes a weighted energy-delay cost per UAV under practical flight constraints (finite horizontal range, collision avoidance, and an elevation-angle limit) and resource constraints. We then develop a mean-field actor-critic (MFAC) framework that aggregates neighbors’ influence into a mean action and conditions both the actor and the critic on local observations and the mean action. By approximating the interactions among a large number of agents through aggregating the influence of others into a mean action representation, the input dimensionality of the critic part is reduced from $M+KP$ to $M+2P$ , yielding an approximately K-fold reduction and becoming independent of the agent population size compared to traditional MADRL methods. Numerical results demonstrate that our proposed algorithm can achieve an 80% reduction in the number of episodes, a 70% reduction in training time, a 38% reduction in energy consumption and a 28% reduction in task delay compared to state-of-the-art approaches, particularly under large-scale UAV deployment scenarios.
{"title":"Task Offloading and Position Optimization for Large-Scale Unmanned Aerial Vehicle Networks: A Mean Field Learning Approach","authors":"Huixian Gu;Liqiang Zhao;Kai Liang;Gan Zheng;Kai-Kit Wong;Chan-Byoung Chae","doi":"10.1109/TCCN.2025.3641515","DOIUrl":"10.1109/TCCN.2025.3641515","url":null,"abstract":"Unmanned aerial vehicle (UAV) networks have emerged as promising enablers in sixth generation (6G) communication system because they can support delay-sensitive and energy-constrained applications. However, the limited resources of UAVs and the high computational complexity of traditional methods complicate task offloading and position optimization. At scale, the task offloading and position optimization decisions yield non-stationary interactions among many agents, while standard multi-agent deep reinforcement learning (MADRL) suffers from poor scalability as the joint action space grows exponentially with the number of UAVs. We formulate joint task offloading and 2D position control as a Markov game that minimizes a weighted energy-delay cost per UAV under practical flight constraints (finite horizontal range, collision avoidance, and an elevation-angle limit) and resource constraints. We then develop a mean-field actor-critic (MFAC) framework that aggregates neighbors’ influence into a mean action and conditions both the actor and the critic on local observations and the mean action. By approximating the interactions among a large number of agents through aggregating the influence of others into a mean action representation, the input dimensionality of the critic part is reduced from <inline-formula> <tex-math>$M+KP$ </tex-math></inline-formula> to <inline-formula> <tex-math>$M+2P$ </tex-math></inline-formula>, yielding an approximately K-fold reduction and becoming independent of the agent population size compared to traditional MADRL methods. Numerical results demonstrate that our proposed algorithm can achieve an 80% reduction in the number of episodes, a 70% reduction in training time, a 38% reduction in energy consumption and a 28% reduction in task delay compared to state-of-the-art approaches, particularly under large-scale UAV deployment scenarios.","PeriodicalId":13069,"journal":{"name":"IEEE Transactions on Cognitive Communications and Networking","volume":"12 ","pages":"4502-4516"},"PeriodicalIF":7.0,"publicationDate":"2025-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145704017","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}
{"title":"Phase-aware Signal Detector for Accurate Classification and Time-Frequency Localization in Wideband Spectrogram","authors":"Chunhui Li, Xin Xiang, Yuan Liang, Qiao Li, Siting Lv, Pengyu Dong","doi":"10.1109/tccn.2025.3641523","DOIUrl":"https://doi.org/10.1109/tccn.2025.3641523","url":null,"abstract":"","PeriodicalId":13069,"journal":{"name":"IEEE Transactions on Cognitive Communications and Networking","volume":"22 1","pages":""},"PeriodicalIF":8.6,"publicationDate":"2025-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145704072","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}
{"title":"Knowledge-Driven 3D Semantic Spectrum Map: KE-VQ-Transformer Based UAV Semantic Communication and Map Completion","authors":"Wei Wu, Lingyi Wang, Fuhui Zhou, Zhaohui Yang, Qihui Wu","doi":"10.1109/tccn.2025.3641514","DOIUrl":"https://doi.org/10.1109/tccn.2025.3641514","url":null,"abstract":"","PeriodicalId":13069,"journal":{"name":"IEEE Transactions on Cognitive Communications and Networking","volume":"11 1","pages":""},"PeriodicalIF":8.6,"publicationDate":"2025-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145704076","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}
Accurate radio map (RM) construction is essential to enabling environment-aware and adaptive wireless communication. However, in future 6G scenarios characterized by high-speed network entities and fast-changing environments, it is very challenging to meet real-time requirements. Although generative diffusion models (DMs) can achieve state-of-the-art accuracy with second-level delay, their iterative nature leads to prohibitive inference latency in delay-sensitive scenarios. In this paper, by uncovering a key structural property of diffusion processes: the latent midpoints remain highly consistent across semantically similar scenes, we propose RadioDiff-Flux, a novel two-stage latent diffusion framework that decouples static environmental modeling from dynamic refinement, enabling the reuse of precomputed midpoints to bypass redundant denoising. In particular, the first stage generates a coarse latent representation using only static scene features, which can be cached and shared across similar scenarios. The second stage adapts this representation to dynamic conditions and transmitter locations using a pre-trained model, thereby avoiding repeated early-stage computation. The proposed RadioDiff-Flux significantly reduces inference time while preserving fidelity. Experiment results show that RadioDiff-Flux can achieve up to $50times $ acceleration with less than 0.15% accuracy loss, demonstrating its practical utility for fast, scalable RM generation in future 6G networks.
{"title":"RadioDiff-Flux: Efficient Radio Map Construction via Generative Denoise Diffusion Model Trajectory Midpoint Reuse","authors":"Xiucheng Wang;Peilin Zheng;Honggang Jia;Nan Cheng;Ruijin Sun;Conghao Zhou;Xuemin Shen","doi":"10.1109/TCCN.2025.3641513","DOIUrl":"10.1109/TCCN.2025.3641513","url":null,"abstract":"Accurate radio map (RM) construction is essential to enabling environment-aware and adaptive wireless communication. However, in future 6G scenarios characterized by high-speed network entities and fast-changing environments, it is very challenging to meet real-time requirements. Although generative diffusion models (DMs) can achieve state-of-the-art accuracy with second-level delay, their iterative nature leads to prohibitive inference latency in delay-sensitive scenarios. In this paper, by uncovering a key structural property of diffusion processes: the latent midpoints remain highly consistent across semantically similar scenes, we propose RadioDiff-Flux, a novel two-stage latent diffusion framework that decouples static environmental modeling from dynamic refinement, enabling the reuse of precomputed midpoints to bypass redundant denoising. In particular, the first stage generates a coarse latent representation using only static scene features, which can be cached and shared across similar scenarios. The second stage adapts this representation to dynamic conditions and transmitter locations using a pre-trained model, thereby avoiding repeated early-stage computation. The proposed RadioDiff-Flux significantly reduces inference time while preserving fidelity. Experiment results show that RadioDiff-Flux can achieve up to <inline-formula> <tex-math>$50times $ </tex-math></inline-formula> acceleration with less than 0.15% accuracy loss, demonstrating its practical utility for fast, scalable RM generation in future 6G networks.","PeriodicalId":13069,"journal":{"name":"IEEE Transactions on Cognitive Communications and Networking","volume":"12 ","pages":"4882-4895"},"PeriodicalIF":7.0,"publicationDate":"2025-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145704079","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}
Pub Date : 2025-12-08DOI: 10.1109/TCCN.2025.3641512
Jianing Wang;Tianhao Liu;Tao Zhang;Jian Wang;Xuangou Wu;Jiqiang Liu
With the rapid development of intelligent transportation and low-altitude economy, collaborative distillation has emerged as an efficient distributed learning paradigm that enables multi-terminal cooperation without sharing raw data. However, the open participation nature of such systems also exposes them to label-flipping attacks, where adversarial devices intentionally alter the labels of their local datasets to mislead the global model aggregation, thus compromising the reliability and trustworthiness of collaborative intelligence. To address this challenge, this paper proposes a label-flipping attack defense framework based on temporal feature modeling and unsupervised anomaly detection. In the early global training phase, each device’s multi-round prediction results on a shared public dataset are collected to extract statistical descriptors, including class distribution, prediction entropy, and inter-class transition strength—forming a short-term temporal feature sequence. An autoencoder is then employed to learn the evolution pattern of benign devices, and the reconstruction error combined with statistical deviation is used to compute anomaly scores. These scores are further analyzed by Isolation Forest model to identify potential attackers in an unsupervised manner. Finally, the anomaly scores are mapped to trust weights for weighted aggregation during subsequent distillation rounds, dynamically suppressing the influence of malicious devices. The proposed framework requires no access to raw data or model parameters, achieving accurate and robust attack detection while preserving privacy. Experimental results demonstrate that the method effectively identifies label-flipping devices under varying attack intensities, significantly enhancing the reliability and robustness of collaborative distillation systems, and showing strong potential for deployment in secure and scalable collaborative learning scenarios.
{"title":"An Adaptive Consistency-Based Detection Scheme for Label-Flipping Attacks in Low-Altitude Economic Networks","authors":"Jianing Wang;Tianhao Liu;Tao Zhang;Jian Wang;Xuangou Wu;Jiqiang Liu","doi":"10.1109/TCCN.2025.3641512","DOIUrl":"10.1109/TCCN.2025.3641512","url":null,"abstract":"With the rapid development of intelligent transportation and low-altitude economy, collaborative distillation has emerged as an efficient distributed learning paradigm that enables multi-terminal cooperation without sharing raw data. However, the open participation nature of such systems also exposes them to label-flipping attacks, where adversarial devices intentionally alter the labels of their local datasets to mislead the global model aggregation, thus compromising the reliability and trustworthiness of collaborative intelligence. To address this challenge, this paper proposes a label-flipping attack defense framework based on temporal feature modeling and unsupervised anomaly detection. In the early global training phase, each device’s multi-round prediction results on a shared public dataset are collected to extract statistical descriptors, including class distribution, prediction entropy, and inter-class transition strength—forming a short-term temporal feature sequence. An autoencoder is then employed to learn the evolution pattern of benign devices, and the reconstruction error combined with statistical deviation is used to compute anomaly scores. These scores are further analyzed by Isolation Forest model to identify potential attackers in an unsupervised manner. Finally, the anomaly scores are mapped to trust weights for weighted aggregation during subsequent distillation rounds, dynamically suppressing the influence of malicious devices. The proposed framework requires no access to raw data or model parameters, achieving accurate and robust attack detection while preserving privacy. Experimental results demonstrate that the method effectively identifies label-flipping devices under varying attack intensities, significantly enhancing the reliability and robustness of collaborative distillation systems, and showing strong potential for deployment in secure and scalable collaborative learning scenarios.","PeriodicalId":13069,"journal":{"name":"IEEE Transactions on Cognitive Communications and Networking","volume":"12 ","pages":"4839-4849"},"PeriodicalIF":7.0,"publicationDate":"2025-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145704071","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}
Pub Date : 2025-12-08DOI: 10.1109/TCCN.2025.3641588
Jiahao You;Ziye Jia;Can Cui;Chao Dong;Qihui Wu;Zhu Han
The low-altitude intelligent networks (LAINs) emerge as a promising architecture for delivering low-latency and energy-efficient edge intelligence in dynamic and infrastructure-limited environments. By integrating uncrewed aerial vehicles (UAVs), aerial base stations, and terrestrial base stations, LAINs can support mission-critical applications such as disaster response, environmental monitoring, and real-time sensing. However, these systems face key challenges, including energy-constrained UAVs, stochastic task arrivals, and heterogeneous computing resources. To address these issues, we propose an integrated air-ground collaborative network and formulate a time-dependent integer nonlinear programming problem that jointly optimizes UAV trajectory planning and task offloading decisions. The problem is challenging to solve due to temporal coupling among decision variables. Therefore, we design a hierarchical learning framework with two timescales. At the large timescale, a Vickrey-Clarke-Groves auction mechanism enables the energy-aware and incentive-compatible trajectory assignment. At the small timescale, we propose the diffusion-heterogeneous-agent proximal policy optimization, a generative multi-agent reinforcement learning algorithm that embeds latent diffusion models into actor networks. Each UAV samples actions from a Gaussian prior and refines them via observation-conditioned denoising, enhancing adaptability and policy diversity. Extensive simulations show that our framework outperforms baselines in energy efficiency, task success rate, and convergence performance.
{"title":"Hierarchical Task Offloading and Trajectory Optimization in Low-Altitude Intelligent Networks via Auction and Diffusion-Based MARL","authors":"Jiahao You;Ziye Jia;Can Cui;Chao Dong;Qihui Wu;Zhu Han","doi":"10.1109/TCCN.2025.3641588","DOIUrl":"10.1109/TCCN.2025.3641588","url":null,"abstract":"The low-altitude intelligent networks (LAINs) emerge as a promising architecture for delivering low-latency and energy-efficient edge intelligence in dynamic and infrastructure-limited environments. By integrating uncrewed aerial vehicles (UAVs), aerial base stations, and terrestrial base stations, LAINs can support mission-critical applications such as disaster response, environmental monitoring, and real-time sensing. However, these systems face key challenges, including energy-constrained UAVs, stochastic task arrivals, and heterogeneous computing resources. To address these issues, we propose an integrated air-ground collaborative network and formulate a time-dependent integer nonlinear programming problem that jointly optimizes UAV trajectory planning and task offloading decisions. The problem is challenging to solve due to temporal coupling among decision variables. Therefore, we design a hierarchical learning framework with two timescales. At the large timescale, a Vickrey-Clarke-Groves auction mechanism enables the energy-aware and incentive-compatible trajectory assignment. At the small timescale, we propose the diffusion-heterogeneous-agent proximal policy optimization, a generative multi-agent reinforcement learning algorithm that embeds latent diffusion models into actor networks. Each UAV samples actions from a Gaussian prior and refines them via observation-conditioned denoising, enhancing adaptability and policy diversity. Extensive simulations show that our framework outperforms baselines in energy efficiency, task success rate, and convergence performance.","PeriodicalId":13069,"journal":{"name":"IEEE Transactions on Cognitive Communications and Networking","volume":"12 ","pages":"5161-5175"},"PeriodicalIF":7.0,"publicationDate":"2025-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145704073","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：