IEEE Transactions on Mobile Computing最新文献_第5页

Smart Multi-Scenario Task Deployment for AUV Cluster Network: A Large Language Model-Driven Exploration-Enhanced MARL Approach AUV集群网络的智能多场景任务部署：一种大型语言模型驱动的探索增强MARL方法

IF 8.8 2区计算机科学

IEEE Transactions on Mobile Computing Pub Date : 2026-07-01 Epub Date: 2026-02-13 DOI: 10.1109/TMC.2026.3664900

Shengchao Zhu;Guangjie Han;Chuan Lin;Chuanliang Chen;Zhenyu Wang;Fan Yang;Tongwei Zhang

{"title":"Smart Multi-Scenario Task Deployment for AUV Cluster Network: A Large Language Model-Driven Exploration-Enhanced MARL Approach","authors":"Shengchao Zhu;Guangjie Han;Chuan Lin;Chuanliang Chen;Zhenyu Wang;Fan Yang;Tongwei Zhang","doi":"10.1109/TMC.2026.3664900","DOIUrl":"https://doi.org/10.1109/TMC.2026.3664900","url":null,"abstract":"Recent advances in network technologies and Multi-Agent Reinforcement Learning (MARL) have accelerated the development of Autonomous Underwater Vehicle (AUV) cluster networks, enabling intelligent applications such as target tracking and cooperative target encirclement. However, existing MARL models are typically designed for single-task scenarios, limiting their scalability in real-world multi-task environments. To address this, we propose a lightweight MARL framework capable of handling multiple AUV tasks with reduced reliance on underwater sampling. Specifically, a unified state space representation is constructed to support task generalization, while a hybrid online-offline MARL training paradigm is introduced by leveraging the logical reasoning and sample generation capabilities of Large Language Models (LLMs). This reduces the demand for real-time data collection. Furthermore, a supervised pretraining strategy is incorporated to improve convergence and learning stability. Based on these components, we develop the <bold>Large <bold>Language <bold>Model-driven <bold>Hybrid online-offline MARL algorithm towards <bold>Multi-<bold>Task scenarios (LLM-HMT), which supports intelligent deployment of multi-task AUV cluster systems with minimal state representation, reduced sample requirements, and limited training iterations. Extensive experiments demonstrate that LLM-HMT outperforms mainstream MARL baselines in convergence speed, task success rate, and resource efficiency, highlighting its potential for practical underwater applications.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"25 7","pages":"10507-10524"},"PeriodicalIF":8.8,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148518416","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

C-PRISM: A Comprehensive Privacy-Preserving and Behavioral-Incentive Mechanism for Sustainable Mobile Crowdsensing C-PRISM：可持续移动众测的综合隐私保护和行为激励机制

IF 8.8 2区计算机科学

IEEE Transactions on Mobile Computing Pub Date : 2026-07-01 Epub Date: 2026-02-16 DOI: 10.1109/TMC.2026.3665505

Jing Bai;Mande Xie;Houbing Song;Mianxiong Dong;Tian Wang;Anfeng Liu

{"title":"C-PRISM: A Comprehensive Privacy-Preserving and Behavioral-Incentive Mechanism for Sustainable Mobile Crowdsensing","authors":"Jing Bai;Mande Xie;Houbing Song;Mianxiong Dong;Tian Wang;Anfeng Liu","doi":"10.1109/TMC.2026.3665505","DOIUrl":"https://doi.org/10.1109/TMC.2026.3665505","url":null,"abstract":"Mobile Crowdsensing (MCS) has emerged as a promising paradigm for large-scale, real-time data collection by leveraging the sensing capabilities of widely distributed mobile workers. However, its practical adoption is challenged by privacy risks and unsustainable incentive structures that inadequately compensate for workers’ inherent participation costs, leading to diminished motivation, sparse task coverage, and reduced data availability. Existing approaches either provide limited and utility-degrading privacy protection or design incentive mechanisms that incur substantial costs, even under the Nash equilibrium. To bridge this gap, we propose C-PRISM (Comprehensive Privacy-preserving and Behavioral-Incentive Sustainable crowdsensing Mechanism), an integrated framework that seamlessly combines privacy-preserving techniques with behavioral-economic incentive design. Specifically, C-PRISM employs randomized matrix perturbation for fine-grained location protection and a two-phase proxy re-encryption protocol to secure task details and sensing data across evaluation, recruitment, and transmission. Building upon this secure foundation, behavioral-economic incentives grounded in prospect theory, the Aronson effect, and a dual reference-point model are introduced to promote sustained worker participation at sub-Nash-equilibrium costs. Rigorous theoretical analysis validates C-PRISM’s security and individual rationality. Extensive experiments on real-world datasets demonstrate that C-PRISM increases data collection efficiency by 7.42%-193.75%, improves worker retention by 2.53%-69.78%, and effectively maintains overall system utility.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"25 7","pages":"10619-10635"},"PeriodicalIF":8.8,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148518417","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

RAMACoDrive: A Real-Time Asynchronous Framework for Cooperative Perception With Realistic V2X Communication RAMACoDrive：一种具有逼真V2X通信的实时异步协作感知框架

IF 8.8 2区计算机科学

IEEE Transactions on Mobile Computing Pub Date : 2026-07-01 Epub Date: 2026-02-16 DOI: 10.1109/TMC.2026.3664934

Shunyao Zhang;Yingfeng Cai;Chengzhi Gao;Caizhen He;Hai Wang;Long Chen

{"title":"RAMACoDrive: A Real-Time Asynchronous Framework for Cooperative Perception With Realistic V2X Communication","authors":"Shunyao Zhang;Yingfeng Cai;Chengzhi Gao;Caizhen He;Hai Wang;Long Chen","doi":"10.1109/TMC.2026.3664934","DOIUrl":"https://doi.org/10.1109/TMC.2026.3664934","url":null,"abstract":"Recent advances in cooperative driving have garnered significant attention, as it is believed to have greater potential than single-vehicle systems in complex environments. However, gaps in current research frameworks impede its further development. Dataset-based frameworks provide a limited perspective, failing to capture the dynamic, asynchronous, and interactive complexities of the real world. Simulation-based frameworks often employ synchronous implementation and oversimplified communication models, impeding meaningful real-time performance evaluation in asynchronous environments. To this end, we propose RAMACoDrive, a novel real-time asynchronous framework for cooperative perception research. RAMACoDrive implements complete process separation, allowing agents and their internal components to operate asynchronously and independently. It uniquely integrates hardware-in-the-loop V2X communication using an external device for realistic data forwarding. Furthermore, RAMACoDrive incorporates an asynchronous online evaluation method with real-time performance metrics, enabling comprehensive evaluation beyond mere accuracy. Through experiments centered on cooperative 3D object detection, we demonstrate its effectiveness in revealing the impact of real-world asynchronous effects and communication constraints. RAMACoDrive provides a valuable and realistic testbed for cooperative perception algorithms and helps to advance the deployment of practical applications. The code is available on our GitHub repository.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"25 7","pages":"10525-10541"},"PeriodicalIF":8.8,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148519020","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Adaptive RIS Configuration Design With Environmental Sensing for User Localization in Dynamic Rich Scattering Environment 基于环境感知的动态富散射环境下用户定位自适应RIS配置设计

IF 8.8 2区计算机科学

IEEE Transactions on Mobile Computing Pub Date : 2026-07-01 Epub Date: 2026-01-21 DOI: 10.1109/TMC.2026.3656736

Anum Umer;Ivo Müürsepp;Muhammad Mahtab Alam

{"title":"Adaptive RIS Configuration Design With Environmental Sensing for User Localization in Dynamic Rich Scattering Environment","authors":"Anum Umer;Ivo Müürsepp;Muhammad Mahtab Alam","doi":"10.1109/TMC.2026.3656736","DOIUrl":"https://doi.org/10.1109/TMC.2026.3656736","url":null,"abstract":"This paper addresses the problem of adaptive reconfigurable intelligent surfaces (RIS) configuration design for user localization in rich-scattering environment (RSE), where electromagnetic waves undergomultiple interactions with dynamic scatterers and RIS elements. We propose an adaptive learning-based localization approach for a distributed RIS-assisted network in a RSE using a bidirectional long-short term memory (biLSTM) model that captures temporal correlations between observations. The proposed approach actively senses the environment using sequential pilot transmissions from the base station (BS), accounting for scattering effects, and adaptively updates the RIS configuration based on prior measurements to eventually accurately estimate and minimize the user localization error. The proposed model comprises two neural sub-networks: Scattering Estimation Network (Bi-SEN), for estimation of scattering in the environment, and Adaptive RIS-Assisted User Localization Network (Bi-ARULN), for RIS configuration and localization. Bayesian optimization is used for hyperparameter tuning of the model. The simulation results demonstrate the effectiveness of the proposed approach, achieving significantly lower localization root mean squared error (RMSE) compared to random configuration, prestored codebook look-ups, and adaptive baselines in both single-input-single-output (SISO) and multiple-input–multiple-output (MIMO) RIS-assisted networks in RSE. The design is generalized across configurations and scales with RIS size and network dimensions. The results highlight the strong potential of RIS deployment and of the proposed approach to enable reliable location services in RSE.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"25 7","pages":"9770-9783"},"PeriodicalIF":8.8,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148519022","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Hierarchical Constraint Fusion for Robust 3D Rigid Body Localization Using UWB Networks 基于超宽带网络的分层约束融合鲁棒三维刚体定位

IF 8.8 2区计算机科学

IEEE Transactions on Mobile Computing Pub Date : 2026-07-01 Epub Date: 2026-01-16 DOI: 10.1109/TMC.2026.3654960

Hongji Yan;You Li;Zixuan Huang;Bingpeng Zhou;Xueli Guo;Xuehang Sun

{"title":"Hierarchical Constraint Fusion for Robust 3D Rigid Body Localization Using UWB Networks","authors":"Hongji Yan;You Li;Zixuan Huang;Bingpeng Zhou;Xueli Guo;Xuehang Sun","doi":"10.1109/TMC.2026.3654960","DOIUrl":"https://doi.org/10.1109/TMC.2026.3654960","url":null,"abstract":"Ultra-wideband rigid body localization (RBL) represents a pivotal technology for achieving high-precision indoor localization. However, it encounters significant challenges, particularly non-line-of-sight (NLOS) propagation errors and reduced vertical observability in complex environments, both of which considerably undermine the robustness of existing three-dimensional(3D) RBL techniques. To mitigate these challenges, this paper introduces a novel, robust 3D-RBL methodology based on hierarchical constraint fusion, referred to as HC-RBL. This approach enhances performance with a three-tier optimization framework: 1) at the sensor layer, the Particle Swarm Optimization algorithm is used to derive the globally optimal geometric configuration of sensor anchor nodes and tag nodes, thereby maximizing spatial observability; 2) at the signal layer, a hybrid robust M-estimation coupled with multiple outlier detection techniques is leveraged to effectively suppress the localization error; 3) at the trajectory layer, the Rauch-Tung-Striebel smoothing algorithm is applied, incorporating rigid-body kinematic constraints to ensure the physical consistency of 3D motion trajectory. The dataset was obtained from indoor wheeled robot experiments, covering approximately 500 m of trajectory over 16 min. In indoor NLOS environments, with a ranging accuracy of 0.31 m, experimental results from multiple wheeled robot localization tests demonstrate that HC-RBL significantly outperforms conventional RBL methods, achieving an 76% reduction in root mean square error to 0.103 m and an 80% reduction in the 95% cumulative error to 0.182 m. The proposed HC-RBL method exhibits remarkable robustness in complex indoor environments against pronounced NLOS effects.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"25 7","pages":"9928-9943"},"PeriodicalIF":8.8,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148515679","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

EAT: QoS-Aware Edge-Collaborative AIGC Task Scheduling via Attention-Guided Diffusion Reinforcement Learning 基于注意引导扩散强化学习的qos感知边缘协同AIGC任务调度

IF 8.8 2区计算机科学

IEEE Transactions on Mobile Computing Pub Date : 2026-07-01 Epub Date: 2026-01-21 DOI: 10.1109/TMC.2026.3656318

Zhifei Xu;Zhiqing Tang;Jiong Lou;Zhi Yao;Xuan Xie;Tian Wang;Yinglong Wang;Weijia Jia

{"title":"EAT: QoS-Aware Edge-Collaborative AIGC Task Scheduling via Attention-Guided Diffusion Reinforcement Learning","authors":"Zhifei Xu;Zhiqing Tang;Jiong Lou;Zhi Yao;Xuan Xie;Tian Wang;Yinglong Wang;Weijia Jia","doi":"10.1109/TMC.2026.3656318","DOIUrl":"https://doi.org/10.1109/TMC.2026.3656318","url":null,"abstract":"The growth of Artificial Intelligence (AI) and large language models has enabled the use of Generative AI (GenAI) in cloud data centers for diverse AI-Generated Content (AIGC) tasks. Models like Stable Diffusion introduce unavoidable delays and substantial resource overhead, which are unsuitable for users at the network edge with high QoS demands. Deploying AIGC services on edge servers reduces transmission times but often leads to underutilized resources and fails to optimally balance inference latency and quality. To address these issues, this paper introduces a QoS-aware <underline>Edge-collaborative <underline>AIGC <underline>Task scheduling (EAT) algorithm. Specifically: 1) We segment AIGC tasks and schedule patches to various edge servers, formulating it as a gang scheduling problem that balances inference latency and quality while considering server heterogeneity, such as differing model distributions and cold start issues. 2) We propose a reinforcement learning-based EAT algorithm that uses a cross attention layer to extract load and task queue information from edge servers and employs a diffusion-based policy network for scheduling, efficiently enabling model reuse. 3) We develop an AIGC task scheduling system that uses our EAT algorithm to divide tasks and distribute them across multiple edge servers for processing. Experimental results based on our system and large-scale simulations show that our EAT algorithm reduces the response time by up to 42% compared to baselines.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"25 7","pages":"9980-9996"},"PeriodicalIF":8.8,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148515680","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Space-Time Block Codec Based Cooperative Integrated Sensing and Communication System 基于空时块编解码器的协同集成传感与通信系统

IF 8.8 2区计算机科学

IEEE Transactions on Mobile Computing Pub Date : 2026-07-01 Epub Date: 2026-01-20 DOI: 10.1109/TMC.2026.3655733

Lin Wang;Zhiyong Feng;Zhiqing Wei;Xinyi Wang;Dingyou Ma;Zesong Fei

{"title":"Space-Time Block Codec Based Cooperative Integrated Sensing and Communication System","authors":"Lin Wang;Zhiyong Feng;Zhiqing Wei;Xinyi Wang;Dingyou Ma;Zesong Fei","doi":"10.1109/TMC.2026.3655733","DOIUrl":"https://doi.org/10.1109/TMC.2026.3655733","url":null,"abstract":"Uncrewed aerial vehicles (UAVs) are poised for explosive growth in the low-altitude economy, causing spectrum congestion and posing a challenge to airspace regulation. Although integrated sensing and communication (ISAC) enables simultaneous communication and sensing, alleviating the spectrum shortage, the capability of one single base station (BS) is generally limited. Therefore, a multi-BS cooperative ISAC system is developed to perceive the status of UAVs at the cell edge. Multiple BSs share the same time-frequency resources and adopt a time-division scheme to avoid mutual interference between communication and sensing functionalities. Specifically, the frame structure of the communication system is modified to accommodate the sensing functionality. A robust interference nulling based beam pattern is first proposed to prevent the line-of-sight (LoS) interference between BSs from overrunning the dynamic range of the analog-to-digital converter (ADC). Moreover, we designed a space-time block codec-based orthogonal frequency division multiplexing (OFDM) to separate echo signals originating from different BSs, which transforms the inter-BS reflected interference into bistatic sensing signals. Furthermore, a data-level fusion method based on the signal-to-interference-plus-noise ratio (SINR) of the range profile is applied to improve the positioning accuracy. The numerical results reveal that the proposed beam pattern greatly avoids LoS interference. The echo signals originating from neighboring BSs can assist in target detection and angle of arrival (AoA) estimation. Compared to soft fusion and single-BS schemes, the proposed fusion method enhances positioning precision by an order of magnitude, and is practically feasible even in the presence of clock synchronization errors.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"25 7","pages":"9366-9381"},"PeriodicalIF":8.8,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148517110","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

AceNet: Attention-Guided Context Enhancement for Imbalanced Action Recognition via RF Signals 基于射频信号的不平衡动作识别的注意引导上下文增强

IF 8.8 2区计算机科学

IEEE Transactions on Mobile Computing Pub Date : 2026-07-01 Epub Date: 2026-02-23 DOI: 10.1109/TMC.2026.3666114

Biyun Sheng;Hao Liu;Hui Cai;Yiping Zuo;Jian Zhou;Fu Xiao

{"title":"AceNet: Attention-Guided Context Enhancement for Imbalanced Action Recognition via RF Signals","authors":"Biyun Sheng;Hao Liu;Hui Cai;Yiping Zuo;Jian Zhou;Fu Xiao","doi":"10.1109/TMC.2026.3666114","DOIUrl":"https://doi.org/10.1109/TMC.2026.3666114","url":null,"abstract":"Although radio frequency (RF)-based activity recognition has made significant progress in recent years, the sensing performance will be significantly degraded under class imbalance conditions, especially when minority and majority classes share semantically similar local motion patterns. Traditional data augmentation approaches in the original sample space may cause semantic deviation and meanwhile bring high computational cost. Instead, in this work we turn to address the issue at feature level, in which we focus on how to distinguish highly similar actions and mitigate imbalance-induced decision boundary bias. To tackle these challenges, we present attention-guided context enhancement network (AceNet), which designs a discriminative feature extractor and develops a feature-augmentation based classifier refinement strategy. Specifically, an attention-guided mechanism is presented to dynamically select the most distinctive temporal segments, and a hierarchical Transformer structure is then proposed to characterize both inner-segment micro-dynamics and inter-segment contextual relationships. Moreover, AceNet synthesizes features via Synthetic Minority Over-sampling Technique (SMOTE) to balance feature distribution for each category and then refine the classifier parameters to mitigate class imbalance bias. Comprehensive experiments on two public datasets with different RF modalities demonstrate that AceNet significantly outperforms existing approaches under various levels of data scarcity at a low cost.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"25 7","pages":"10455-10470"},"PeriodicalIF":8.8,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148517106","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

A Model Consistency-Based Countermeasure to GAN-Based Data Poisoning Attack in Federated Learning 基于模型一致性的联邦学习中基于gan的数据中毒攻击对策

IF 8.8 2区计算机科学

IEEE Transactions on Mobile Computing Pub Date : 2026-07-01 Epub Date: 2026-02-23 DOI: 10.1109/TMC.2026.3665336

Wei Sun;Bo Gao;Ke Xiong;Yuwei Wang;Pingyi Fan;Khaled Ben Letaief

{"title":"A Model Consistency-Based Countermeasure to GAN-Based Data Poisoning Attack in Federated Learning","authors":"Wei Sun;Bo Gao;Ke Xiong;Yuwei Wang;Pingyi Fan;Khaled Ben Letaief","doi":"10.1109/TMC.2026.3665336","DOIUrl":"https://doi.org/10.1109/TMC.2026.3665336","url":null,"abstract":"In federated learning (FL), although the original intention of “available but not visible” data is to allay data privacy concerns, it potentially brings new security threats, particularly poisoning attacks that target such “not visible” local data. Intuitively, such data poisoning attacks have great potential in stealthily degrading global FL outcomes, and are expected to be even stealthier if being enhanced by generative models like generative adversarial networks (GANs). However, existing defense methods have not been thoroughly challenged in this regard and generally fail to be aware of a local generation of seemingly legitimate poisoned data. With a growing concern on potentially stealthier attacks, in this paper, a cost-effective defense mechanism named Model Consistency-Based Defense (MCD) is proposed, which offers a comprehensive examination of available local models across multiple feature dimensions, providing an indirect yet effective means of identifying hidden data poisoning attackers. To push the limit of MCD against stealthier attacks, we propose a new GAN-based data poisoning attack model named VagueGAN and an unsupervised variant of it, which can be flexibly deployed to generate seemingly legitimate but noisy poisoned data. The consistency of GAN outputs revealed by VagueGAN helps strengthen MCD to work against stealthier GAN-based attacks as well as other mainstream ones. Extensive experiments on multiple open datasets (MNIST, Fashion-MNIST, CIFAR-10, CIFAR-100, and Mini-Imagenet) indicate that our attack method better balances the trade-off between attack effectiveness and stealthiness with low complexity. More importantly, our defense mechanism is shown to be more competent in identifying a variety of poisoned data, particularly stealthier GAN-poisoned ones.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"25 7","pages":"10351-10369"},"PeriodicalIF":8.8,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148517465","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

DRL-Based Wireless-Powered UAVs Trajectories Planning for Fair Communication 基于drl的无线动力无人机公平通信轨迹规划

IF 8.8 2区计算机科学

IEEE Transactions on Mobile Computing Pub Date : 2026-07-01 Epub Date: 2026-02-12 DOI: 10.1109/TMC.2026.3664292

Peixiang Wang;Xiaoyu Wang;He Huang;Haipeng Dai

{"title":"DRL-Based Wireless-Powered UAVs Trajectories Planning for Fair Communication","authors":"Peixiang Wang;Xiaoyu Wang;He Huang;Haipeng Dai","doi":"10.1109/TMC.2026.3664292","DOIUrl":"https://doi.org/10.1109/TMC.2026.3664292","url":null,"abstract":"While unmanned aerial vehicle base stations (UAV-BSs) offer transformative potential for enhancing terrestrial networks, their deployment faces dual challenges of limited onboard energy and dynamic channel conditions that compromise long-term fair communication coverage. In this paper, we investigate the problem of <bold>Wireless-powered <bold>UAVs <bold>Trajectories planning for <bold>Fair communication (WUTF), that is, navigating multiple UAV-BSs powered by wireless charging towers (WCTs), to provide fair communication services for ground users. To address the problem, we first formulate the original optimization problem as a Partially Observable Markov Decision Process (POMDP), and propose a Deep Reinforcement Learning (DRL) based trajectory planning algorithm. The proposed approach incorporates a novel reward function that balances multiple objectives and adopts a Centralized Training with Executed Decentralization (CTED) framework. Furthermore, a sequential policy update scheme is introduced to enhance multi-UAV coordination and reduce policy conflicts. Simulation results show that our proposed algorithm significantly improves the communication fairness, total throughput, and communication efficiency, as compared to state-of-the-art DRL-based methods up to 32.01% on average.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"25 7","pages":"10317-10334"},"PeriodicalIF":8.8,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148517464","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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