Pub Date : 2025-09-15DOI: 10.1109/TMC.2025.3609967
Yanyan Wang;Jia Liu;Zhihao Qu;Shen-Huan Lyu;Bin Tang;Baoliu Ye
With the proliferation of RFID-enabled applications, large-scale RFID systems often require multiple readers to ensure full coverage of numerous tags. In such systems, we sometimes pay more attention to a subset of tags instead of all, which are called key tags. This paper studies an under-investigated problem key tag distribution identification, which aims to identify which key tags are beneath which readers. This is crucial for efficiently managing specific items of interest, which can quickly pinpoint key tags and help RFID readers covering these tags collaborate to improve the tag inventory efficiency. We propose a protocol called Kadept that identifies the key tag distribution by designing a sophisticated Cuckoo filter that teases out key tags as well as assigns each of them a singleton slot for response. With this design, a great number of trivial (non-key) tags will keep silent and free up bandwidth resources for key tags, and each key tag is sorted in a collision-free way and can be identified with only 1-bit response, which significantly improves the time efficiency. To enhance the scalability and efficiency of Kadept for high key tag proportions, we propose E-Kadept protocol, which accelerates the identification process by designing an incremental Cuckoo filter that reduces false positives and improves space efficiency. We theoretically analyze how to optimize protocol parameters of Kadept and E-Kadept, and conduct extensive simulations under different tag distribution scenarios. Compared with the state-of-the-art, E-Kadept can improve the time efficiency by a factor of 1.75×, when the ratio of key tags to all tags is 0.3.
{"title":"Time-Efficient Identifying Key Tag Distribution in Large-Scale RFID Systems","authors":"Yanyan Wang;Jia Liu;Zhihao Qu;Shen-Huan Lyu;Bin Tang;Baoliu Ye","doi":"10.1109/TMC.2025.3609967","DOIUrl":"https://doi.org/10.1109/TMC.2025.3609967","url":null,"abstract":"With the proliferation of RFID-enabled applications, large-scale RFID systems often require multiple readers to ensure full coverage of numerous tags. In such systems, we sometimes pay more attention to a subset of tags instead of all, which are called key tags. This paper studies an under-investigated problem <italic>key tag distribution identification</i>, which aims to identify which key tags are beneath which readers. This is crucial for efficiently managing specific items of interest, which can quickly pinpoint key tags and help RFID readers covering these tags collaborate to improve the tag inventory efficiency. We propose a protocol called Kadept that identifies the key tag distribution by designing a sophisticated Cuckoo filter that teases out key tags as well as assigns each of them a singleton slot for response. With this design, a great number of trivial (non-key) tags will keep silent and free up bandwidth resources for key tags, and each key tag is sorted in a collision-free way and can be identified with only 1-bit response, which significantly improves the time efficiency. To enhance the scalability and efficiency of Kadept for high key tag proportions, we propose E-Kadept protocol, which accelerates the identification process by designing an incremental Cuckoo filter that reduces false positives and improves space efficiency. We theoretically analyze how to optimize protocol parameters of Kadept and E-Kadept, and conduct extensive simulations under different tag distribution scenarios. Compared with the state-of-the-art, E-Kadept can improve the time efficiency by a factor of 1.75×, when the ratio of key tags to all tags is 0.3.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"25 2","pages":"2725-2742"},"PeriodicalIF":9.2,"publicationDate":"2025-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145929394","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}
Pub Date : 2025-09-12DOI: 10.1109/TMC.2025.3609316
Shiming Yu;Ziyue Zhang;Xianjin Xia;Yuanqing Zheng;Jiliang Wang
LoRaWANs are envisioned to connect billions of IoT devices through thousands of physically overlapping yet logically orthogonal channels (termed logical channels). These logical channels hold significant potential for enabling highly concurrent scalable IoT connectivity. Large-scale deployments however face strong interference between logical channels. This practical issue has been largely overlooked by existing works but becomes increasingly prominent as LoRaWAN scales up. To address this issue, we introduce Canas, an innovative gateway design that is poised to orthogonalize the logical channels by eliminating mutual interference. To this end, Canas develops a series of novel solutions to accurately extract the meta-information of individual ultra-weak LoRa signals from the received overlapping channels. The meta-information is then leveraged to accurately reconstruct and subtract the LoRa signals over thousands of logical channels iteratively. Real-world evaluations demonstrate that Canas can enhance concurrent transmissions across overlapping logical channels by 2.3× compared to the best known related works.
{"title":"Resolving Inter-Logical Channel Interference for Large-Scale LoRa Deployments","authors":"Shiming Yu;Ziyue Zhang;Xianjin Xia;Yuanqing Zheng;Jiliang Wang","doi":"10.1109/TMC.2025.3609316","DOIUrl":"https://doi.org/10.1109/TMC.2025.3609316","url":null,"abstract":"LoRaWANs are envisioned to connect billions of IoT devices through thousands of physically overlapping yet logically orthogonal channels (termed logical channels). These logical channels hold significant potential for enabling highly concurrent scalable IoT connectivity. Large-scale deployments however face strong interference between logical channels. This practical issue has been largely overlooked by existing works but becomes increasingly prominent as LoRaWAN scales up. To address this issue, we introduce Canas, an innovative gateway design that is poised to orthogonalize the logical channels by eliminating mutual interference. To this end, Canas develops a series of novel solutions to accurately extract the meta-information of individual ultra-weak LoRa signals from the received overlapping channels. The meta-information is then leveraged to accurately reconstruct and subtract the LoRa signals over thousands of logical channels iteratively. Real-world evaluations demonstrate that Canas can enhance concurrent transmissions across overlapping logical channels by 2.3× compared to the best known related works.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"25 2","pages":"2759-2773"},"PeriodicalIF":9.2,"publicationDate":"2025-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145929339","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}
Magnetoquasistatic (MQS) field positioning has demonstrated significant potential for emergency rescue applications due to its strong penetration and non-reliance on pre-deployment. However, its accuracy is notably impaired by metal interference and distance attenuation. Inertial Measurement Units (IMUs) can reliably provide motion data even in environments affected by metal and electromagnetic interference, but they suffer from cumulative drift over time. Effectively, combining MQS field and IMU positioning to harness their respective advantages presents a crucial challenge. To address this, we propose a Multi-Layer Position-Pose Fusion (MP2F) framework that integrates MQS field with IMU data to enhance position and pose estimation. The MP2F framework comprises three layers: a Quaternion-based Pose Fusion Layer (QPFL), a Kalman Filter-based Position Fusion Layer (KFFL), and a Global Position-Pose Fusion Layer (GP2FL). Specifically, QPFL utilizes the Extended Kalman Filter (EKF) to effectively mitigate magnetic field distortion and IMU drift, thereby significantly enhancing pose estimation precision. Next, KFFL incorporates the fused pose estimation from QPFL into an inertial navigation motion model, and leverages MQS field observations to further improve positional accuracy. Finally, GP2FL formulates a nonlinear least squares optimization problem by marginalizing prior factors, inertial sensor factors, and Kalman fusion outputs, enabling globally optimized state estimation. Comprehensive simulation results and analyses prove that the proposed MP2F framework achieves high-precision position and pose estimation in complex emergency scenarios, with strong robustness. Experimental results in real-world environments show that the proposed MP2F achieves improvements in positioning accuracy of 61.1%, 58.7%, 48.4%, and 50.2% over EKF, iMag+, GWO-PF, and MagLoc, respectively.
{"title":"A Multi-Layer Position-Pose Fusion Framework for Joint Magnetoquasistatic Field and IMU Positioning","authors":"Bocheng Qian;Lei Huang;Xiansheng Guo;Gordon Owusu Boateng;Rui Ma;Nirwan Ansari","doi":"10.1109/TMC.2025.3608822","DOIUrl":"https://doi.org/10.1109/TMC.2025.3608822","url":null,"abstract":"Magnetoquasistatic (MQS) field positioning has demonstrated significant potential for emergency rescue applications due to its strong penetration and non-reliance on pre-deployment. However, its accuracy is notably impaired by metal interference and distance attenuation. Inertial Measurement Units (IMUs) can reliably provide motion data even in environments affected by metal and electromagnetic interference, but they suffer from cumulative drift over time. Effectively, combining MQS field and IMU positioning to harness their respective advantages presents a crucial challenge. To address this, we propose a <italic>Multi-Layer Position-Pose Fusion</i> (<italic>MP2F</i>) framework that integrates MQS field with IMU data to enhance position and pose estimation. The MP2F framework comprises three layers: a <italic>Quaternion-based Pose Fusion Layer</i> (<italic>QPFL</i>), a <italic>Kalman Filter-based Position Fusion Layer</i> (<italic>KFFL</i>), and a <italic>Global Position-Pose Fusion Layer</i> (<italic>GP2FL</i>). Specifically, QPFL utilizes the Extended Kalman Filter (EKF) to effectively mitigate magnetic field distortion and IMU drift, thereby significantly enhancing pose estimation precision. Next, KFFL incorporates the fused pose estimation from QPFL into an inertial navigation motion model, and leverages MQS field observations to further improve positional accuracy. Finally, GP2FL formulates a nonlinear least squares optimization problem by marginalizing prior factors, inertial sensor factors, and Kalman fusion outputs, enabling globally optimized state estimation. Comprehensive simulation results and analyses prove that the proposed MP2F framework achieves high-precision position and pose estimation in complex emergency scenarios, with strong robustness. Experimental results in real-world environments show that the proposed MP2F achieves improvements in positioning accuracy of 61.1%, 58.7%, 48.4%, and 50.2% over EKF, iMag+, GWO-PF, and MagLoc, respectively.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"25 2","pages":"2843-2859"},"PeriodicalIF":9.2,"publicationDate":"2025-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145929570","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}
Pub Date : 2025-09-12DOI: 10.1109/TMC.2025.3609480
Ngangbam Indrason;Kalyan Baital;Goutam Saha
The electoral system is one of the fundamental pillars of democracy, but the traditional voting system suffers from several limitations such as fraud voting, vote tampering, impersonation, and inefficiencies. To overcome these limitations, several research works have been initiated to design a blockchain-based e-voting system. These designs addressed the loopholes of the existing ones to a limited extent. Here, a novel multi-level blockchain-secured SDN-based IoT enabled e-voting system has been proposed. The proposed system consists of booth, district, state, and country level systems. Here, a voter needs to be authenticated at the booth-level and then this valid vote data can be propagated to the upper hierarchical levels and stored there after signing and encrypting it using ECDSA and ECC respectively. Man-in-the-middle attacks, DoS/DDoS, unauthorized access, and impersonation attacks are avoided using flow rules in SDN controllers and firewalls installed in the servers. Furthermore, blockchain technology provides security for voting data stored at all levels. The security strengths were tested at different levels (e.g., programming, operating system, and network level) using open-source tools (i.e., scyther, nmap, metasploit, etc.). The performance of the proposed architecture was evaluated satisfactorily in a testbed. It also performed satisfactorily under both normal and stressed conditions in a scaled-up environment.
{"title":"Design and Security Analysis of SDN-Based IoT-Oriented Blockchain Protected E-Voting System","authors":"Ngangbam Indrason;Kalyan Baital;Goutam Saha","doi":"10.1109/TMC.2025.3609480","DOIUrl":"https://doi.org/10.1109/TMC.2025.3609480","url":null,"abstract":"The electoral system is one of the fundamental pillars of democracy, but the traditional voting system suffers from several limitations such as fraud voting, vote tampering, impersonation, and inefficiencies. To overcome these limitations, several research works have been initiated to design a blockchain-based e-voting system. These designs addressed the loopholes of the existing ones to a limited extent. Here, a novel multi-level blockchain-secured SDN-based IoT enabled e-voting system has been proposed. The proposed system consists of booth, district, state, and country level systems. Here, a voter needs to be authenticated at the booth-level and then this valid vote data can be propagated to the upper hierarchical levels and stored there after signing and encrypting it using ECDSA and ECC respectively. Man-in-the-middle attacks, DoS/DDoS, unauthorized access, and impersonation attacks are avoided using flow rules in SDN controllers and firewalls installed in the servers. Furthermore, blockchain technology provides security for voting data stored at all levels. The security strengths were tested at different levels (e.g., programming, operating system, and network level) using open-source tools (i.e., scyther, nmap, metasploit, etc.). The performance of the proposed architecture was evaluated satisfactorily in a testbed. It also performed satisfactorily under both normal and stressed conditions in a scaled-up environment.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"25 2","pages":"2811-2824"},"PeriodicalIF":9.2,"publicationDate":"2025-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145929589","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}
Pub Date : 2025-09-10DOI: 10.1109/TMC.2025.3608245
Dawei Yan;Feiyu Han;Mingzhu Yang;Shanyue Wang;Panlong Yang;Yubo Yan
Currently, a major issue of WiFi-based sensing technologies is how to adapt to changes in the surrounding environment. The extreme sensitivity of Channel State Information (CSI) makes many WiFi sensing arts frustrated when applied to the complex and unknown real world. To solve this problem, in this paper, we propose freeEnv designed to automatically identify the micro-environmental changes (even tiny movements of the laptop) using WiFi devices, which can coexist with other WiFi sensing tasks with zero effort. To achieve automatic identification of micro-environmental changes, we quantify micro-environmental changes based on the physical propagation laws of WiFi signals and the main factors that affect CSI measurements. Then, we design a micro-environmental changes identification method, which determines whether the environment has changed by calculating the Earth Mover’s Distance (EMD) of the Probability Density Function (PDF) of continuous CSI, without requiring training data. To remove the influence of dynamic human behaviors, we design a human dynamic detection scheme, which is achieved by obtaining the average inter-cluster distance of performing Gaussian Mixture Model (GMM) clustering on CSI. We evaluate freeEnv in real-world scenarios with six different hardware, four different scenarios, and twenty-four ways of micro-environmental changes. The results show that our method is robust to different devices and scenarios, and can achieve the average precision of 96.1% and 93.2% for micro-environmental changes identification and human dynamic behavior detection. By testing on a case study of threshold-based human presence detection, freeEnv can effectively improve the detection performance.
{"title":"freeEnv: Enabling Zero-Effort RF-Based Micro-Environment Changes Monitoring","authors":"Dawei Yan;Feiyu Han;Mingzhu Yang;Shanyue Wang;Panlong Yang;Yubo Yan","doi":"10.1109/TMC.2025.3608245","DOIUrl":"https://doi.org/10.1109/TMC.2025.3608245","url":null,"abstract":"Currently, a major issue of WiFi-based sensing technologies is how to adapt to changes in the surrounding environment. The extreme sensitivity of <italic>Channel State Information</i> (CSI) makes many WiFi sensing arts frustrated when applied to the complex and unknown real world. To solve this problem, in this paper, we propose <italic>freeEnv</i> designed to automatically identify the micro-environmental changes (even tiny movements of the laptop) using WiFi devices, which can coexist with other WiFi sensing tasks with zero effort. To achieve automatic identification of micro-environmental changes, we quantify micro-environmental changes based on the physical propagation laws of WiFi signals and the main factors that affect CSI measurements. Then, we design a micro-environmental changes identification method, which determines whether the environment has changed by calculating the <italic>Earth Mover’s Distance</i> (EMD) of the <italic>Probability Density Function</i> (PDF) of continuous CSI, without requiring training data. To remove the influence of dynamic human behaviors, we design a human dynamic detection scheme, which is achieved by obtaining the average inter-cluster distance of performing <italic>Gaussian Mixture Model</i> (GMM) clustering on CSI. We evaluate <italic>freeEnv</i> in real-world scenarios with six different hardware, four different scenarios, and twenty-four ways of micro-environmental changes. The results show that our method is robust to different devices and scenarios, and can achieve the average precision of 96.1% and 93.2% for micro-environmental changes identification and human dynamic behavior detection. By testing on a case study of threshold-based human presence detection, <italic>freeEnv</i> can effectively improve the detection performance.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"25 2","pages":"2743-2758"},"PeriodicalIF":9.2,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145929395","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}
Pub Date : 2025-09-10DOI: 10.1109/TMC.2025.3608303
Xinyi Xu;Gang Feng;Yijing Liu;Shuang Qin;Jian Wang;Yunxiang Wang
Large Language Models (LLMs), with advanced content creation and inference capabilities, can provide immersive intelligent services to users in mobile edge networks. However, the increasing demand for real-time artificial intelligence (AI) applications aggravates the limitations of cloud-based LLMs due to the long response time. Meanwhile, Small Language Models (SLMs), which are cost-effective and locally deployable for terminal devices, can serve as an efficient supplement to LLMs for performing latency-sensitive tasks with lower generalization capability. Due to the resource constraints of edge networks and the diverse requirements of user tasks, it is critical to design an inference framework that effectively coordinates the deployment and collaboration of LLMs and SLMs. In this paper, we propose an LLM-SLM collaborative inference (LSCI) scheme under a mobile edge computing (MEC) architecture, which jointly decides where to cache models and how to offload inference tasks to balance latency, accuracy, and resource costs. To optimize inference performance subject to resource constraints, we jointly solve the inference task offloading and model caching problem in LSCI scheme. Specifically, we employ deep reinforcement learning (DRL) to select highly popular SLMs to be cached on the edge server, and distributed belief propagation technique to solve the associated inference task offloading issue. Numerical results show that the proposed LSCI scheme can achieve significant performance gain in terms of inference performance when compared with a number of baseline solutions.
大型语言模型（llm）具有先进的内容创建和推理能力，可以为移动边缘网络中的用户提供沉浸式智能服务。然而，对实时人工智能（AI）应用的需求不断增长，由于响应时间长，加剧了基于云的法学硕士的局限性。同时，小型语言模型（Small Language Models, slm）具有成本效益和可在终端设备本地部署的特点，可以作为llm的有效补充，用于执行延迟敏感型任务，但泛化能力较低。由于边缘网络的资源限制和用户任务的多样化需求，设计一个有效协调llm和slm部署和协作的推理框架至关重要。在本文中，我们提出了一种移动边缘计算（MEC）架构下的LLM-SLM协同推理（LSCI）方案，该方案共同决定在何处缓存模型以及如何卸载推理任务，以平衡延迟、准确性和资源成本。为了优化资源约束下的推理性能，我们共同解决了LSCI方案中的推理任务卸载和模型缓存问题。具体而言，我们采用深度强化学习（DRL）选择高度流行的slm缓存到边缘服务器，并采用分布式信念传播技术解决相关的推理任务卸载问题。数值结果表明，与许多基准方案相比，所提出的LSCI方案在推理性能方面取得了显著的性能提升。
{"title":"Joint Inference Offloading and Model Caching for Small and Large Language Model Collaboration","authors":"Xinyi Xu;Gang Feng;Yijing Liu;Shuang Qin;Jian Wang;Yunxiang Wang","doi":"10.1109/TMC.2025.3608303","DOIUrl":"https://doi.org/10.1109/TMC.2025.3608303","url":null,"abstract":"Large Language Models (LLMs), with advanced content creation and inference capabilities, can provide immersive intelligent services to users in mobile edge networks. However, the increasing demand for real-time artificial intelligence (AI) applications aggravates the limitations of cloud-based LLMs due to the long response time. Meanwhile, Small Language Models (SLMs), which are cost-effective and locally deployable for terminal devices, can serve as an efficient supplement to LLMs for performing latency-sensitive tasks with lower generalization capability. Due to the resource constraints of edge networks and the diverse requirements of user tasks, it is critical to design an inference framework that effectively coordinates the deployment and collaboration of LLMs and SLMs. In this paper, we propose an LLM-SLM collaborative inference (LSCI) scheme under a mobile edge computing (MEC) architecture, which jointly decides where to cache models and how to offload inference tasks to balance latency, accuracy, and resource costs. To optimize inference performance subject to resource constraints, we jointly solve the inference task offloading and model caching problem in LSCI scheme. Specifically, we employ deep reinforcement learning (DRL) to select highly popular SLMs to be cached on the edge server, and distributed belief propagation technique to solve the associated inference task offloading issue. Numerical results show that the proposed LSCI scheme can achieve significant performance gain in terms of inference performance when compared with a number of baseline solutions.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"25 2","pages":"2691-2706"},"PeriodicalIF":9.2,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145929506","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}
Nowadays IoT devices in Mobile Edge Computing (MEC) networks have been deployed in large-scale quantities to guarantee sensing data collection for anomalous event detection as full as possible even if some devices are in fault. Some techniques, such as clustering and dimensionality reduction, are adopted to eliminate redundant sensing data collection in this large-scale deployment. However, they not only have high computational complexity and easily cause the loss of information on the primary sensing attributes for detection, but also bring certain errors to the detection because of their low sensitivity to data processed. In addition, insufficient collection of primary attribute data samples often results from physical or human factors, and mindless imputation of large-scale data gaps without basis may lead to greater irreparable losses. To address the above challenges, we first complete the selection of optimal primary attribute device collection and aggregation (PADCA) path based on minimum spanning tree, reducing data communication cost for redundant primary attributes collection. Then, we propose an anomalous impact correlation search strategy to quickly locate all MEC servers whose management regions have cascading anomalous event and help determine the transferable source MEC servers. Leveraging this, we use transfer learning to help detect anomalous events in the management regions of the MEC servers with insufficient primary attribute data samples, where a particle swarm optimization based back-propagation (PSO-BP) neural network model is used to infer the fusion weight of each primary attribute. Experimental results show that our method achieves higher detection performance in terms of detection time, energy consumption, accuracy, and receiver operating characteristic (ROC) curve compared to the benchmarks by at least 24%, 34%, 0.5 and 0.05.
{"title":"Transfer Learning Assisted Detection of Anomalous Events With Insufficient Primary Attribute Data Samples in MEC Networks","authors":"Jine Tang;Xiaotong Ma;Song Yang;Yong Xiang;Zhangbing Zhou","doi":"10.1109/TMC.2025.3604253","DOIUrl":"https://doi.org/10.1109/TMC.2025.3604253","url":null,"abstract":"Nowadays IoT devices in Mobile Edge Computing (MEC) networks have been deployed in large-scale quantities to guarantee sensing data collection for anomalous event detection as full as possible even if some devices are in fault. Some techniques, such as clustering and dimensionality reduction, are adopted to eliminate redundant sensing data collection in this large-scale deployment. However, they not only have high computational complexity and easily cause the loss of information on the primary sensing attributes for detection, but also bring certain errors to the detection because of their low sensitivity to data processed. In addition, insufficient collection of primary attribute data samples often results from physical or human factors, and mindless imputation of large-scale data gaps without basis may lead to greater irreparable losses. To address the above challenges, we first complete the selection of optimal primary attribute device collection and aggregation (PADCA) path based on minimum spanning tree, reducing data communication cost for redundant primary attributes collection. Then, we propose an anomalous impact correlation search strategy to quickly locate all MEC servers whose management regions have cascading anomalous event and help determine the transferable source MEC servers. Leveraging this, we use transfer learning to help detect anomalous events in the management regions of the MEC servers with insufficient primary attribute data samples, where a particle swarm optimization based back-propagation (PSO-BP) neural network model is used to infer the fusion weight of each primary attribute. Experimental results show that our method achieves higher detection performance in terms of detection time, energy consumption, accuracy, and receiver operating characteristic (ROC) curve compared to the benchmarks by at least 24%, 34%, 0.5 and 0.05.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"25 1","pages":"1254-1269"},"PeriodicalIF":9.2,"publicationDate":"2025-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145659211","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}
The rapid developments of mobile devices and fog computing have facilitated the data collection paradigm of fog-assisted mobile crowdsensing, providing great convenience for individuals with limited resources to collect large-scale data. However, the openness of crowdsensing network and the untrusted behaviors of some task participants raise concerns regarding participants’ privacy and data reliability. Previous works mostly focus on preserving the privacy of task participants and often overlook the issue of data reliability in the presence of dishonest participants. In this paper, we propose a new data collection scheme tailored for fog-assisted mobile crowdsensing. It enables the cloud to detect invalid sensing data in the ciphertext domain, simultaneously ensuring data confidentiality and reliability. Additionally, our scheme is designed to protect the anonymity of honest task participants while guaranteeing the traceability of dishonest participant once invalid data are detected. Formal analysis is provided to prove the correctness and security of our scheme. Furthermore, we implement our scheme to evaluate its performance, and the experimental results demonstrate that it can achieve the aforementioned security properties with modest performance overhead.
{"title":"Enabling Reliable and Anonymous Data Collection for Fog-Assisted Mobile Crowdsensing With Malicious User Detection","authors":"Mingyang Song;Zhongyun Hua;Yifeng Zheng;Rushi Lan;Qing Liao;Guoai Xu","doi":"10.1109/TMC.2025.3602659","DOIUrl":"https://doi.org/10.1109/TMC.2025.3602659","url":null,"abstract":"The rapid developments of mobile devices and fog computing have facilitated the data collection paradigm of fog-assisted mobile crowdsensing, providing great convenience for individuals with limited resources to collect large-scale data. However, the openness of crowdsensing network and the untrusted behaviors of some task participants raise concerns regarding participants’ privacy and data reliability. Previous works mostly focus on preserving the privacy of task participants and often overlook the issue of data reliability in the presence of dishonest participants. In this paper, we propose a new data collection scheme tailored for fog-assisted mobile crowdsensing. It enables the cloud to detect invalid sensing data in the ciphertext domain, simultaneously ensuring data confidentiality and reliability. Additionally, our scheme is designed to protect the anonymity of honest task participants while guaranteeing the traceability of dishonest participant once invalid data are detected. Formal analysis is provided to prove the correctness and security of our scheme. Furthermore, we implement our scheme to evaluate its performance, and the experimental results demonstrate that it can achieve the aforementioned security properties with modest performance overhead.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"25 1","pages":"1414-1430"},"PeriodicalIF":9.2,"publicationDate":"2025-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145659222","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}
The evolution of industrial intelligent manufacturing necessitates wireless communication systems capable of replacing conventional wired infrastructures, offering superior flexibility, scalability, and reduced maintenance overhead. While 5G New Radio (NR) Ultra-Reliable Low-Latency Communication (uRLLC) standards (Release 15-17) have shown promise for mission-critical applications, current implementations remain constrained by their unidirectional optimization paradigm, unable to simultaneously satisfy the dual imperatives of sub-millisecond latency ($< 1$ ms) and 99.9999% reliability demanded by industrial control systems. To address these challenges, we present a transformative subband full-duplex (SBFD) network architecture that ensures persistent time-domain spectral availability for concurrent uplink/downlink operations, thereby eliminating direction-switching latency. Our solution introduces three key innovations: (1) an interference-aware SBFD resource allocation framework that strategically isolates UL/DL subbands to minimize cross-link interference (CLI), (2) a dual-optimization algorithm that jointly maximizes spectral efficiency while guaranteeing channel-adaptive reliability thresholds, and (3) a practical implementation scheme compatible with existing 5G NR physical layer specifications. Extensive simulations under realistic factory channel models demonstrate 58.3% reduction in aggregate CLI and 41.2% improvement in control command decoding accuracy compared to legacy half-duplex systems. This research establishes a new paradigm for wireless industrial networks, effectively closing the performance gap between 5G URLLC specifications and the exacting demands of Industry 4.0 applications.
{"title":"Uplink and Downlink Subband Resource Allocation for Subband Full-Duplex Enabled Industrial Intelligent Manufacturing","authors":"Zheng Jiang;Dingyou Ma;Bowen Wang;Ningyan Guo;Kan Yu;Qixun Zhang","doi":"10.1109/TMC.2025.3602872","DOIUrl":"https://doi.org/10.1109/TMC.2025.3602872","url":null,"abstract":"The evolution of industrial intelligent manufacturing necessitates wireless communication systems capable of replacing conventional wired infrastructures, offering superior flexibility, scalability, and reduced maintenance overhead. While 5G New Radio (NR) Ultra-Reliable Low-Latency Communication (uRLLC) standards (Release 15-17) have shown promise for mission-critical applications, current implementations remain constrained by their unidirectional optimization paradigm, unable to simultaneously satisfy the dual imperatives of sub-millisecond latency (<inline-formula><tex-math>$< 1$</tex-math></inline-formula> ms) and 99.9999% reliability demanded by industrial control systems. To address these challenges, we present a transformative subband full-duplex (SBFD) network architecture that ensures persistent time-domain spectral availability for concurrent uplink/downlink operations, thereby eliminating direction-switching latency. Our solution introduces three key innovations: (1) an interference-aware SBFD resource allocation framework that strategically isolates UL/DL subbands to minimize cross-link interference (CLI), (2) a dual-optimization algorithm that jointly maximizes spectral efficiency while guaranteeing channel-adaptive reliability thresholds, and (3) a practical implementation scheme compatible with existing 5G NR physical layer specifications. Extensive simulations under realistic factory channel models demonstrate 58.3% reduction in aggregate CLI and 41.2% improvement in control command decoding accuracy compared to legacy half-duplex systems. This research establishes a new paradigm for wireless industrial networks, effectively closing the performance gap between 5G URLLC specifications and the exacting demands of Industry 4.0 applications.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"25 1","pages":"1431-1444"},"PeriodicalIF":9.2,"publicationDate":"2025-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145659192","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}
Pub Date : 2025-08-26DOI: 10.1109/TMC.2025.3603092
Mainak Chakraborty;Chandan;Bodhibrata Mukhopadhyay;Subrat Kar
In this paper, we introduce a framework for multi-class novelty detection using structural vibration signals. Structural vibration-based person identification is a promising soft-biometric approach with potential applications in elderly care and access control. However, current research faces two key challenges. The first challenge is the lack of large-scale datasets necessary for thorough evaluation in structural vibration gait recognition. To address this, we created a new dataset with recordings from fifty individuals. The second challenge lies in the limited exploration of deep learning methods for large-scale multi-class novelty detection in structural vibration data. To fill this gap, we propose the energy-shifted contrastive loss function, specifically designed for this task. Our results demonstrate that the proposed framework achieves 96.57% accuracy in multi-class classification. For novelty detection, it achieves an Receiver Operating Characteristic—Area Under the Curve (ROC-AUC) score of 89.15% for single footsteps, which improves to 93.83% with five consecutive footsteps.
本文介绍了一种利用结构振动信号进行多类新颖性检测的框架。基于结构振动的人识别是一种很有前途的软生物识别方法，在老年人护理和门禁控制方面具有潜在的应用前景。然而，目前的研究面临两个关键挑战。第一个挑战是缺乏对结构振动步态识别进行全面评估所需的大规模数据集。为了解决这个问题，我们创建了一个包含50个人录音的新数据集。第二个挑战是深度学习方法在结构振动数据中大规模多类新颖性检测方面的探索有限。为了填补这一空白，我们提出了能量转移对比损失函数，专门为这项任务设计。结果表明，该框架在多类分类中准确率达到96.57%。在新颖性检测方面，单步行走的ROC-AUC （Receiver Operating Characteristic-Area Under the Curve）得分为89.15%，连续5步行走的ROC-AUC得分为93.83%。
{"title":"Deep Multi-Class Novelty Detection in Structural Vibrations With Modified Contrastive Loss","authors":"Mainak Chakraborty;Chandan;Bodhibrata Mukhopadhyay;Subrat Kar","doi":"10.1109/TMC.2025.3603092","DOIUrl":"https://doi.org/10.1109/TMC.2025.3603092","url":null,"abstract":"In this paper, we introduce a framework for multi-class novelty detection using structural vibration signals. Structural vibration-based person identification is a promising soft-biometric approach with potential applications in elderly care and access control. However, current research faces two key challenges. The first challenge is the lack of large-scale datasets necessary for thorough evaluation in structural vibration gait recognition. To address this, we created a new dataset with recordings from fifty individuals. The second challenge lies in the limited exploration of deep learning methods for large-scale multi-class novelty detection in structural vibration data. To fill this gap, we propose the energy-shifted contrastive loss function, specifically designed for this task. Our results demonstrate that the proposed framework achieves 96.57% accuracy in multi-class classification. For novelty detection, it achieves an Receiver Operating Characteristic—Area Under the Curve (ROC-AUC) score of 89.15% for single footsteps, which improves to 93.83% with five consecutive footsteps.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"25 1","pages":"1457-1468"},"PeriodicalIF":9.2,"publicationDate":"2025-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145659216","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}
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：