Pub Date : 2025-02-05DOI: 10.1109/TMC.2025.3527174
{"title":"2024 Reviewers List","authors":"","doi":"10.1109/TMC.2025.3527174","DOIUrl":"https://doi.org/10.1109/TMC.2025.3527174","url":null,"abstract":"","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"24 3","pages":"2470-2484"},"PeriodicalIF":7.7,"publicationDate":"2025-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10874877","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143361325","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-12-02DOI: 10.1109/TMC.2024.3499958
Meng Zhang;Deepanshu Vasal
Network utility maximization (NUM) is a fundamental framework for optimizing next-generation networks. However, self-interested agents with private information pose challenges due to potential system manipulation. To address these challenges, the literature on economic mechanism design has emerged. Existing mechanisms are not suited for large-scale networks due to their complexity, high implementation costs, and difficulty to adapt to dynamic settings. This paper proposes a large-scale mechanism design framework that mitigates these limitations. As the number of agents $I$ approaches infinity, their incentive to misreport decreases rapidly at a rate of $mathcal {O}(1/I^{2})$. We introduce a superimposable framework applicable to any NUM algorithm without modifications, reducing implementation costs. In the dynamic setting, the large-scale mechanism design framework introduces the decomposability of the problem, enabling agents to align their own interests with the objectives of the dynamic NUM problem. This alignment helps overcome the additional, more stringent incentive constraints encountered in dynamic settings. Extending our results to dynamic settings, we present the design of a Dynamic Large-Scale mechanism with desirable properties and the corresponding Dynamic Superimposable Large-Scale mechanism. Our numerical experiments validate the fact that our proposed schemes are approximately $I$ times faster than the seminal VCG mechanism.
网络效用最大化（NUM）是优化下一代网络的基本框架。然而，由于潜在的系统操纵，拥有私人信息的自利代理带来了挑战。为了应对这些挑战，有关经济机制设计的文献应运而生。现有的机制因其复杂性、高实施成本和难以适应动态设置而不适合大规模网络。本文提出的大规模机制设计框架可减轻这些限制。当代理的数量 $I$ 接近无穷大时，他们错误报告的动机会以 $mathcal {O}(1/I^{2})$ 的速度迅速降低。我们引入了一个可叠加框架，无需修改即可适用于任何 NUM 算法，从而降低了实施成本。在动态环境中，大规模机制设计框架引入了问题的可分解性，使代理能够将自身利益与动态 NUM 问题的目标相一致。这种协调有助于克服动态环境中遇到的额外、更严格的激励约束。将我们的结果扩展到动态环境中，我们提出了具有理想特性的动态大规模机制设计以及相应的动态可叠加大规模机制。我们的数值实验验证了我们提出的方案比开创性的 VCG 机制快约 $I$ 倍。
{"title":"Large-Scale Mechanism Design for Networks: Superimposability and Dynamic Implementation","authors":"Meng Zhang;Deepanshu Vasal","doi":"10.1109/TMC.2024.3499958","DOIUrl":"https://doi.org/10.1109/TMC.2024.3499958","url":null,"abstract":"Network utility maximization (NUM) is a fundamental framework for optimizing next-generation networks. However, self-interested agents with private information pose challenges due to potential system manipulation. To address these challenges, the literature on economic mechanism design has emerged. Existing mechanisms are not suited for large-scale networks due to their complexity, high implementation costs, and difficulty to adapt to dynamic settings. This paper proposes a large-scale mechanism design framework that mitigates these limitations. As the number of agents <inline-formula><tex-math>$I$</tex-math></inline-formula> approaches infinity, their incentive to misreport decreases rapidly at a rate of <inline-formula><tex-math>$mathcal {O}(1/I^{2})$</tex-math></inline-formula>. We introduce a superimposable framework applicable to any NUM algorithm without modifications, reducing implementation costs. In the dynamic setting, the large-scale mechanism design framework introduces the decomposability of the problem, enabling agents to align their own interests with the objectives of the dynamic NUM problem. This alignment helps overcome the additional, more stringent incentive constraints encountered in dynamic settings. Extending our results to dynamic settings, we present the design of a Dynamic Large-Scale mechanism with desirable properties and the corresponding Dynamic Superimposable Large-Scale mechanism. Our numerical experiments validate the fact that our proposed schemes are approximately <inline-formula><tex-math>$I$</tex-math></inline-formula> times faster than the seminal VCG mechanism.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"24 3","pages":"1278-1292"},"PeriodicalIF":7.7,"publicationDate":"2024-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143184148","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 : 2024-11-25DOI: 10.1109/TMC.2024.3506161
Yuhang Wang;Ying He;F. Richard Yu;Kaishun Wu;Shanzhi Chen
Intelligent transportation systems demand efficient resource allocation and task offloading to ensure low-latency, high-bandwidth vehicular services. The dynamic nature of vehicular environments, characterized by high mobility and extensive interactions among vehicles, necessitates considering time-varying statistical regularities, especially in scenarios with sharp variations. Despite the widespread use of traditional reinforcement learning for resource allocation, its limitations in generalization and interpretability are evident. To overcome these challenges, we propose an Intelligence-based Reinforcement Learning (IRL) algorithm. This algorithm utilizes active inference to infer the real world and maintain an internal model by minimizing free energy. Enhancing the efficiency of active inference, we incorporate prior knowledge as macro guidance, ensuring more accurate and efficient training. By constructing an intelligence-based model, we eliminate the need for designing reward functions, aligning better with human thinking, and providing a method to reflect the learning, information transmission and intelligence accumulation processes. This approach also allows for quantifying intelligence to a certain extent. Considering the dynamic and uncertain nature of vehicular scenarios, we apply the IRL algorithm to environments with constantly changing parameters. Extensive simulations confirm the effectiveness of IRL, significantly improving the generalization and interpretability of intelligent models in vehicular networks.
{"title":"Intelligence-Based Reinforcement Learning for Dynamic Resource Optimization in Edge Computing-Enabled Vehicular Networks","authors":"Yuhang Wang;Ying He;F. Richard Yu;Kaishun Wu;Shanzhi Chen","doi":"10.1109/TMC.2024.3506161","DOIUrl":"https://doi.org/10.1109/TMC.2024.3506161","url":null,"abstract":"Intelligent transportation systems demand efficient resource allocation and task offloading to ensure low-latency, high-bandwidth vehicular services. The dynamic nature of vehicular environments, characterized by high mobility and extensive interactions among vehicles, necessitates considering time-varying statistical regularities, especially in scenarios with sharp variations. Despite the widespread use of traditional reinforcement learning for resource allocation, its limitations in generalization and interpretability are evident. To overcome these challenges, we propose an Intelligence-based Reinforcement Learning (IRL) algorithm. This algorithm utilizes active inference to infer the real world and maintain an internal model by minimizing free energy. Enhancing the efficiency of active inference, we incorporate prior knowledge as macro guidance, ensuring more accurate and efficient training. By constructing an intelligence-based model, we eliminate the need for designing reward functions, aligning better with human thinking, and providing a method to reflect the learning, information transmission and intelligence accumulation processes. This approach also allows for quantifying intelligence to a certain extent. Considering the dynamic and uncertain nature of vehicular scenarios, we apply the IRL algorithm to environments with constantly changing parameters. Extensive simulations confirm the effectiveness of IRL, significantly improving the generalization and interpretability of intelligent models in vehicular networks.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"24 3","pages":"2394-2406"},"PeriodicalIF":7.7,"publicationDate":"2024-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143361010","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 : 2024-11-15DOI: 10.1109/TMC.2024.3499371
Kai Li;Jingjing Zheng;Wei Ni;Hailong Huang;Pietro Liò;Falko Dressler;Ozgur B. Akan
Fairness in Federated Learning (FL) is imperative not only for the ethical utilization of technology but also for ensuring that models provide accurate, equitable, and beneficial outcomes across varied user demographics and equipment. This paper proposes a new adversarial architecture, referred to as Adversarial Graph Attention Network (AGAT), which deliberately instigates fairness attacks with an aim to bias the learning process across the FL. The proposed AGAT is developed to synthesize malicious, biasing model updates, where the minimum of Kullback-Leibler (KL) divergence between the user's model update and the global model is maximized. Due to a limited set of labeled input-output biasing data samples, a surrogate model is created, which presents the behavior of a complex malicious model update. Moreover, a graph autoencoder (GAE) is designed within the AGAT architecture, which is trained together with sub-gradient descent to reconstruct manipulatively the correlations of the model updates, and maximize the reconstruction loss while keeping the malicious, biasing model updates undetectable. The proposed AGAT attack is implemented in PyTorch, showing experimentally that AGAT successfully increases the minimum value of KL divergence of benign model updates by 60.9% and bypasses the detection of existing defense models. The source code of the AGAT attack is released on GitHub.
{"title":"Biasing Federated Learning With a New Adversarial Graph Attention Network","authors":"Kai Li;Jingjing Zheng;Wei Ni;Hailong Huang;Pietro Liò;Falko Dressler;Ozgur B. Akan","doi":"10.1109/TMC.2024.3499371","DOIUrl":"https://doi.org/10.1109/TMC.2024.3499371","url":null,"abstract":"Fairness in Federated Learning (FL) is imperative not only for the ethical utilization of technology but also for ensuring that models provide accurate, equitable, and beneficial outcomes across varied user demographics and equipment. This paper proposes a new adversarial architecture, referred to as Adversarial Graph Attention Network (AGAT), which deliberately instigates fairness attacks with an aim to bias the learning process across the FL. The proposed AGAT is developed to synthesize malicious, biasing model updates, where the minimum of Kullback-Leibler (KL) divergence between the user's model update and the global model is maximized. Due to a limited set of labeled input-output biasing data samples, a surrogate model is created, which presents the behavior of a complex malicious model update. Moreover, a graph autoencoder (GAE) is designed within the AGAT architecture, which is trained together with sub-gradient descent to reconstruct manipulatively the correlations of the model updates, and maximize the reconstruction loss while keeping the malicious, biasing model updates undetectable. The proposed AGAT attack is implemented in PyTorch, showing experimentally that AGAT successfully increases the minimum value of KL divergence of benign model updates by 60.9% and bypasses the detection of existing defense models. The source code of the AGAT attack is released on GitHub.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"24 3","pages":"2407-2421"},"PeriodicalIF":7.7,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143361324","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 unpredictability of the wireless channel has been used as a natural source of randomness to build physical-layer security primitives for shared key generation, authentication, access control, proximity verification, and other security properties. Compared to pseudo-random generators, it has the potential to achieve information-theoretic security. In sub-6 GHz frequencies, the randomness is harvested from the small-scale fading effects of RF signal propagation in rich scattering environments. However, the RF propagation characteristics follow sparse models with clustered paths when devices operate in millimeter-wave (mmWave) bands (5G and Next-Generation networks, Wi-Fi in 60GHz). Millimeter-wave transmissions are typically directional to increase the gain and combat high signal attenuation, leading to stable and more predictable channels. In this paper, we first demonstrate that state-of-the-art methods relying on channel state information or received signal strength measurements fail to produce high randomness. Accounting for the unique features of mmWave propagation, we propose a novel randomness extraction mechanism that exploits the random timing of channel blockage to harvest random bits. Compared with the prior art in CSI-based and context-based randomness extraction, our protocol remains secure against passive and active Man-in-the-Middle adversaries co-located with the legitimate devices. We demonstrate the security properties of our method in a 28 GHz mmWave testbed in an indoor setting.
{"title":"Harvesting Physical-Layer Randomness in Millimeter Wave Bands","authors":"Ziqi Xu;Jingcheng Li;Yanjun Pan;Ming Li;Loukas Lazos","doi":"10.1109/TMC.2024.3499876","DOIUrl":"https://doi.org/10.1109/TMC.2024.3499876","url":null,"abstract":"The unpredictability of the wireless channel has been used as a natural source of randomness to build physical-layer security primitives for shared key generation, authentication, access control, proximity verification, and other security properties. Compared to pseudo-random generators, it has the potential to achieve information-theoretic security. In sub-6 GHz frequencies, the randomness is harvested from the small-scale fading effects of RF signal propagation in rich scattering environments. However, the RF propagation characteristics follow sparse models with clustered paths when devices operate in millimeter-wave (mmWave) bands (5G and Next-Generation networks, Wi-Fi in 60GHz). Millimeter-wave transmissions are typically directional to increase the gain and combat high signal attenuation, leading to stable and more predictable channels. In this paper, we first demonstrate that state-of-the-art methods relying on channel state information or received signal strength measurements fail to produce high randomness. Accounting for the unique features of mmWave propagation, we propose a novel randomness extraction mechanism that exploits the random timing of channel blockage to harvest random bits. Compared with the prior art in CSI-based and context-based randomness extraction, our protocol remains secure against <italic>passive and active Man-in-the-Middle adversaries co-located with the legitimate devices</i>. We demonstrate the security properties of our method in a 28 GHz mmWave testbed in an indoor setting.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"24 3","pages":"2285-2300"},"PeriodicalIF":7.7,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143360989","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 : 2024-11-11DOI: 10.1109/TMC.2024.3496536
Han Li;Ke Xiong;Yuping Lu;Wei Chen;Pingyi Fan;Khaled Ben Letaief
Small-cell mobile edge computing (SE-MEC) networks amalgamate the virtues of MEC and small-cell networks, enhancing data processing capabilities of user devices (UDs). Nevertheless, time-varying wireless channels, dynamic UD requirements, and severe interference among UDs make it difficult to fully exploit the limited network resources and stably provide computing services for UDs. Therefore, efficient task offloading and resource allocation (TORA) is essential. Moreover, since multiple small cells are deployed, decentralized TORA schemes are preferred in practice. Thus, this paper aims to design distributed adaptive TORA schemes for SE-MEC networks. In pursuit of an eco-friendly design, an optimization problem is formulated to minimize the total energy consumption (TEC) of UDs subject to delay constraints. To effectively deal with network's dynamic characteristics, the reinforce learning framework is applied, where the TEC minimization problem is first modeled as a partially observable Markov decision process (POMDP), and then an efficient multi-agent proximal policy optimization (MAPPO)-based scheme is presented to solve it. In the presented scheme, each small-cell base station (SBS) serves as an agent and is capable of making TORA decisions only with its own local information. To promote collaboration among multiple agents, a global reward function is designed. A state normalization mechanism is also introduced into the presented scheme for enhancing learning performance. Simulation results show that although the proposed MAPPO-based scheme works in a distributed manner, it achieves very similar performance to the centralized one. In addition, it is demonstrated that the state normalization mechanism has a significant effect on reducing TEC.
{"title":"Collaborative Task Offloading and Resource Allocation in Small-Cell MEC: A Multi-Agent PPO-Based Scheme","authors":"Han Li;Ke Xiong;Yuping Lu;Wei Chen;Pingyi Fan;Khaled Ben Letaief","doi":"10.1109/TMC.2024.3496536","DOIUrl":"https://doi.org/10.1109/TMC.2024.3496536","url":null,"abstract":"Small-cell mobile edge computing (SE-MEC) networks amalgamate the virtues of MEC and small-cell networks, enhancing data processing capabilities of user devices (UDs). Nevertheless, time-varying wireless channels, dynamic UD requirements, and severe interference among UDs make it difficult to fully exploit the limited network resources and stably provide computing services for UDs. Therefore, efficient task offloading and resource allocation (TORA) is essential. Moreover, since multiple small cells are deployed, decentralized TORA schemes are preferred in practice. Thus, this paper aims to design distributed adaptive TORA schemes for SE-MEC networks. In pursuit of an eco-friendly design, an optimization problem is formulated to minimize the total energy consumption (TEC) of UDs subject to delay constraints. To effectively deal with network's dynamic characteristics, the reinforce learning framework is applied, where the TEC minimization problem is first modeled as a partially observable Markov decision process (POMDP), and then an efficient multi-agent proximal policy optimization (MAPPO)-based scheme is presented to solve it. In the presented scheme, each small-cell base station (SBS) serves as an agent and is capable of making TORA decisions only with its own local information. To promote collaboration among multiple agents, a global reward function is designed. A state normalization mechanism is also introduced into the presented scheme for enhancing learning performance. Simulation results show that although the proposed MAPPO-based scheme works in a distributed manner, it achieves very similar performance to the centralized one. In addition, it is demonstrated that the state normalization mechanism has a significant effect on reducing TEC.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"24 3","pages":"2346-2359"},"PeriodicalIF":7.7,"publicationDate":"2024-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143361009","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}
Various characteristics of mobile applications (apps) and associated in-app services can reveal potentially-sensitive user information; however, privacy concerns have prompted third-party apps to restrict access to data related to mobile app usage. This paper outlines a novel approach to extracting detailed app usage information by analyzing electromagnetic (EM) signals emitted from mobile devices during app-related tasks. The proposed system, MagSpy, recovers user privacy information from magnetometer readings that do not require access permissions. This EM leakage becomes complex when multiple apps are used simultaneously and is subject to interference from geomagnetic signals generated by device movement. To address these challenges, MagSpy employs multiple techniques to extract and identify signals related to app usage. Specifically, the geomagnetic offset signal is canceled using accelerometer and gyroscope sensor data, and a Cascade-LSTM algorithm is used to classify apps and in-app services. MagSpy also uses CWT-based peak detection and a Random Forest classifier to detect PIN inputs. A prototype system was evaluated on over 50 popular mobile apps with 30 devices. Extensive evaluation results demonstrate the efficacy of MagSpy in identifying in-app services (96% accuracy), apps (93.5% accuracy), and extracting PIN input information (96% top-3 accuracy).
{"title":"MagSpy: Revealing User Privacy Leakage via Magnetometer on Mobile Devices","authors":"Yongjian Fu;Lanqing Yang;Hao Pan;Yi-Chao Chen;Guangtao Xue;Ju Ren","doi":"10.1109/TMC.2024.3495506","DOIUrl":"https://doi.org/10.1109/TMC.2024.3495506","url":null,"abstract":"Various characteristics of mobile applications (apps) and associated in-app services can reveal potentially-sensitive user information; however, privacy concerns have prompted third-party apps to restrict access to data related to mobile app usage. This paper outlines a novel approach to extracting detailed app usage information by analyzing electromagnetic (EM) signals emitted from mobile devices during app-related tasks. The proposed system, MagSpy, recovers user privacy information from magnetometer readings that do not require access permissions. This EM leakage becomes complex when multiple apps are used simultaneously and is subject to interference from geomagnetic signals generated by device movement. To address these challenges, MagSpy employs multiple techniques to extract and identify signals related to app usage. Specifically, the geomagnetic offset signal is canceled using accelerometer and gyroscope sensor data, and a Cascade-LSTM algorithm is used to classify apps and in-app services. MagSpy also uses CWT-based peak detection and a Random Forest classifier to detect PIN inputs. A prototype system was evaluated on over 50 popular mobile apps with 30 devices. Extensive evaluation results demonstrate the efficacy of MagSpy in identifying in-app services (96% accuracy), apps (93.5% accuracy), and extracting PIN input information (96% top-3 accuracy).","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"24 3","pages":"2455-2469"},"PeriodicalIF":7.7,"publicationDate":"2024-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143361322","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 : 2024-11-11DOI: 10.1109/TMC.2024.3496443
Baili Chai;Di Wu;Jinyu Chen;Mengyu Yang;Zelong Wang;Miao Hu
The demand for mobile video streaming has seen a substantial surge in recent years. However, current platforms heavily depend on network capacity to ensure the delivery of high-quality video streams. The emergence of neural-enhanced video streaming presents a promising solution to address this challenge by leveraging client-side computation, thereby reducing bandwidth consumption. Nonetheless, deploying advanced super-resolution (SR) models on mobile devices is hindered by the computational demands of existing SR models. In this paper, we propose REM, a novel neural-enhanced mobile video streaming framework. REM utilizes a customized lookup table to facilitate real-time neural-enhanced video streaming on mobile devices. Initially, we conduct a series of measurements to identify abundant macroblock redundancies across frames in a video stream. Subsequently, we introduce a dynamic macroblock selection algorithm that prioritizes important macroblocks for neural enhancement. The SR-enhanced results are stored in the lookup table and efficiently reused to meet real-time requirements and minimize resource overhead. By considering macroblock-level characteristics of the video frames, the lookup table enables efficient and fast processing. Additionally, we design a lightweight macroblock-aware SR module to expedite inference. Finally, we perform extensive experiments on various mobile devices. The results demonstrate that REM enhances overall processing throughput by up to 10.2 times and reduces power consumption by up to 58.6% compared to state-of-the-art methods. Consequently, this leads to a 38.06% improvement in the quality of experience for mobile users.
{"title":"REM: Enabling Real-Time Neural-Enhanced Video Streaming on Mobile Devices Using Macroblock-Aware Lookup Table","authors":"Baili Chai;Di Wu;Jinyu Chen;Mengyu Yang;Zelong Wang;Miao Hu","doi":"10.1109/TMC.2024.3496443","DOIUrl":"https://doi.org/10.1109/TMC.2024.3496443","url":null,"abstract":"The demand for mobile video streaming has seen a substantial surge in recent years. However, current platforms heavily depend on network capacity to ensure the delivery of high-quality video streams. The emergence of neural-enhanced video streaming presents a promising solution to address this challenge by leveraging client-side computation, thereby reducing bandwidth consumption. Nonetheless, deploying advanced super-resolution (SR) models on mobile devices is hindered by the computational demands of existing SR models. In this paper, we propose REM, a novel neural-enhanced mobile video streaming framework. REM utilizes a customized lookup table to facilitate real-time neural-enhanced video streaming on mobile devices. Initially, we conduct a series of measurements to identify abundant macroblock redundancies across frames in a video stream. Subsequently, we introduce a dynamic macroblock selection algorithm that prioritizes important macroblocks for neural enhancement. The SR-enhanced results are stored in the lookup table and efficiently reused to meet real-time requirements and minimize resource overhead. By considering macroblock-level characteristics of the video frames, the lookup table enables efficient and fast processing. Additionally, we design a lightweight macroblock-aware SR module to expedite inference. Finally, we perform extensive experiments on various mobile devices. The results demonstrate that REM enhances overall processing throughput by up to 10.2 times and reduces power consumption by up to 58.6% compared to state-of-the-art methods. Consequently, this leads to a 38.06% improvement in the quality of experience for mobile users.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"24 3","pages":"2085-2097"},"PeriodicalIF":7.7,"publicationDate":"2024-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143360915","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 : 2024-11-11DOI: 10.1109/TMC.2024.3495719
Jingbo Liu;Jiacheng Chen;Zongxi Liu;Haibo Zhou
To enhance flexibility and facilitate resource cooperation, a novel fully-decoupled radio access network (FD-RAN) architecture is proposed for 6G. However, the decoupling of uplink (UL) and downlink (DL) in FD-RAN makes the existing feedback mechanism ineffective. To this end, we propose an end-to-end data-driven MIMO solution without the conventional channel feedback procedure. Data-driven MIMO can alleviate the drawbacks of feedback including overheads and delay, and can provide customized precoding design for different BSs based on their historical channel data. It essentially learns a mapping from geolocation to MIMO transmission parameters. We first present a codebook-based approach, which selects transmission parameters from the statistics of discrete channel state information (CSI) values and utilizes nearest neighbor interpolation for spatial inference. We further present a non-codebook-based approach, which 1) derives the optimal precoder from the singular value decomposition (SVD) of the channel; 2) utilizes variational autoencoder (VAE) to select the representative precoder from the latent Gaussian representations; and 3) exploits Gaussian process regression (GPR) to predict unknown precoders in the space domain. Extensive simulations are performed on a link-level 5G simulator using realistic ray-tracing channel data. The results demonstrate the effectiveness of data-driven MIMO, showcasing its potential for application in FD-RAN and 6G.
{"title":"Enabling Feedback-Free MIMO Transmission for FD-RAN: A Data-Driven Approach","authors":"Jingbo Liu;Jiacheng Chen;Zongxi Liu;Haibo Zhou","doi":"10.1109/TMC.2024.3495719","DOIUrl":"https://doi.org/10.1109/TMC.2024.3495719","url":null,"abstract":"To enhance flexibility and facilitate resource cooperation, a novel fully-decoupled radio access network (FD-RAN) architecture is proposed for 6G. However, the decoupling of uplink (UL) and downlink (DL) in FD-RAN makes the existing feedback mechanism ineffective. To this end, we propose an end-to-end data-driven MIMO solution without the conventional channel feedback procedure. Data-driven MIMO can alleviate the drawbacks of feedback including overheads and delay, and can provide customized precoding design for different BSs based on their historical channel data. It essentially learns a mapping from geolocation to MIMO transmission parameters. We first present a codebook-based approach, which selects transmission parameters from the statistics of discrete channel state information (CSI) values and utilizes nearest neighbor interpolation for spatial inference. We further present a non-codebook-based approach, which 1) derives the optimal precoder from the singular value decomposition (SVD) of the channel; 2) utilizes variational autoencoder (VAE) to select the representative precoder from the latent Gaussian representations; and 3) exploits Gaussian process regression (GPR) to predict unknown precoders in the space domain. Extensive simulations are performed on a link-level 5G simulator using realistic ray-tracing channel data. The results demonstrate the effectiveness of data-driven MIMO, showcasing its potential for application in FD-RAN and 6G.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"24 3","pages":"2437-2454"},"PeriodicalIF":7.7,"publicationDate":"2024-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143361321","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 : 2024-11-08DOI: 10.1109/TMC.2024.3494842
Ning Hui;Qian Sun;Jie Zeng;Lin Tian;Yuanyuan Wang;Yiqing Zhou
Although resource sharing and mixed numerology among slices are promising for improving wireless resource utilization, these techniques can compromise isolation performance and cause serious inter numerology interference (INI). Therefore, this paper studies wireless resource management in a mixed numerology-based sliced 6G space–terrestrial integrated radio access network (STI-RAN) with the aim of reducing INI and guaranteeing isolation performance while decreasing interference from Doppler frequency shifts caused by the high-speed movement of low-orbit satellites. First, an isolation performance indicator is defined to evaluate different isolation performances, and a universal spectral distance model is formulated to rewrite the INI power model. Next, the dynamic wireless resource management problem is formulated in a discrete form, yielding a scheme called Flex-$mu$, which is designed to reduce the INI and Doppler frequency shifts, guarantee isolation performance, and enhance the SINR. Finally, an intelligent multi-characteristic matrix coding-based social group optimization (MultiMatrix-SGO) algorithm is designed to solve the proposed NP-hard discrete optimization problem. Compared with existing schemes, the system utility is efficiently increased by up to 58.32%, the SINR can converge to 38.44 dB, and the isolation performance is guaranteed while the INI and Doppler frequency shifts are reduced.
尽管切片间的资源共享和混合数字技术有望提高无线资源利用率，但这些技术会影响隔离性能，并造成严重的数字间干扰（INI）。因此，本文研究了基于混合数字学的分片 6G 空地一体化无线接入网（STI-RAN）中的无线资源管理，旨在降低 INI，保证隔离性能，同时减少低轨道卫星高速移动造成的多普勒频移干扰。首先，定义了一个隔离性能指标来评估不同的隔离性能，并制定了一个通用频谱距离模型来重写 INI 功率模型。其次，将动态无线资源管理问题离散化，提出了一种名为 Flex-$mu$ 的方案，旨在减少 INI 和多普勒频移，保证隔离性能，提高 SINR。最后，设计了一种基于多特征矩阵编码的智能社会群体优化算法（MultiMatrix-SGO）来解决所提出的 NP-hard离散优化问题。与现有方案相比，系统效用有效提高了 58.32%，信噪比收敛到 38.44 dB，隔离性能得到保证，同时减少了 INI 和多普勒频移。
{"title":"Mixed Numerology-Based Intelligent Resource Management in a Sliced 6G Space–Terrestrial Integrated Radio Access Network","authors":"Ning Hui;Qian Sun;Jie Zeng;Lin Tian;Yuanyuan Wang;Yiqing Zhou","doi":"10.1109/TMC.2024.3494842","DOIUrl":"https://doi.org/10.1109/TMC.2024.3494842","url":null,"abstract":"Although resource sharing and mixed numerology among slices are promising for improving wireless resource utilization, these techniques can compromise isolation performance and cause serious inter numerology interference (INI). Therefore, this paper studies wireless resource management in a mixed numerology-based sliced 6G space–terrestrial integrated radio access network (STI-RAN) with the aim of reducing INI and guaranteeing isolation performance while decreasing interference from Doppler frequency shifts caused by the high-speed movement of low-orbit satellites. First, an isolation performance indicator is defined to evaluate different isolation performances, and a universal spectral distance model is formulated to rewrite the INI power model. Next, the dynamic wireless resource management problem is formulated in a discrete form, yielding a scheme called Flex-<inline-formula><tex-math>$mu$</tex-math></inline-formula>, which is designed to reduce the INI and Doppler frequency shifts, guarantee isolation performance, and enhance the SINR. Finally, an intelligent multi-characteristic matrix coding-based social group optimization (MultiMatrix-SGO) algorithm is designed to solve the proposed NP-hard discrete optimization problem. Compared with existing schemes, the system utility is efficiently increased by up to 58.32%, the SINR can converge to 38.44 dB, and the isolation performance is guaranteed while the INI and Doppler frequency shifts are reduced.","PeriodicalId":50389,"journal":{"name":"IEEE Transactions on Mobile Computing","volume":"24 3","pages":"1338-1356"},"PeriodicalIF":7.7,"publicationDate":"2024-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143184149","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：