Pub Date : 2024-03-21DOI: 10.1109/TMLCN.2024.3403789
Pialy Biswas;Ranjan K. Mallik;Khaled B. Letaief
This paper proposes a Gaussian mixture model (GMM) based access point (AP) clustering technique in cell-free massive MIMO (CFMM) communication systems. The APs are first clustered on the basis of large-scale fading coefficients, and the users are assigned to each cluster depending on the channel gain. As the number of clusters increases, there is a degradation in the overall data rate of the system, causing a trade-off between the cluster number and average rate per user. To address this problem, we present an optimization problem that optimizes both the upper bound on the average downlink rate per user and the number of clusters. The optimal number of clusters is intuitively determined by solving the optimization problem, and then grouping the APs and users. As a result, the computation expense is much lower than the current techniques, since the existing methods require evaluations of the network performance in multiple iterations to find the optimal number of clusters. In addition, we analyze the performance of both balanced and unbalanced clustering. Numerical results will indicate that the unbalanced clustering yields a superior rate per user while maintaining a lower level of complexity compared to the balanced one. Furthermore, we investigate the statistical analysis of the spectral efficiency (SE) per user in the clustered CFMM. The findings reveal that the SE per user can be approximated by the logistic distribution.
{"title":"Optimal Access Point Centric Clustering for Cell-Free Massive MIMO Using Gaussian Mixture Model Clustering","authors":"Pialy Biswas;Ranjan K. Mallik;Khaled B. Letaief","doi":"10.1109/TMLCN.2024.3403789","DOIUrl":"https://doi.org/10.1109/TMLCN.2024.3403789","url":null,"abstract":"This paper proposes a Gaussian mixture model (GMM) based access point (AP) clustering technique in cell-free massive MIMO (CFMM) communication systems. The APs are first clustered on the basis of large-scale fading coefficients, and the users are assigned to each cluster depending on the channel gain. As the number of clusters increases, there is a degradation in the overall data rate of the system, causing a trade-off between the cluster number and average rate per user. To address this problem, we present an optimization problem that optimizes both the upper bound on the average downlink rate per user and the number of clusters. The optimal number of clusters is intuitively determined by solving the optimization problem, and then grouping the APs and users. As a result, the computation expense is much lower than the current techniques, since the existing methods require evaluations of the network performance in multiple iterations to find the optimal number of clusters. In addition, we analyze the performance of both balanced and unbalanced clustering. Numerical results will indicate that the unbalanced clustering yields a superior rate per user while maintaining a lower level of complexity compared to the balanced one. Furthermore, we investigate the statistical analysis of the spectral efficiency (SE) per user in the clustered CFMM. The findings reveal that the SE per user can be approximated by the logistic distribution.","PeriodicalId":100641,"journal":{"name":"IEEE Transactions on Machine Learning in Communications and Networking","volume":"2 ","pages":"675-687"},"PeriodicalIF":0.0,"publicationDate":"2024-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10535986","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141187291","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-20DOI: 10.1109/TMLCN.2024.3403513
Ke Wang;Wanchun Liu;Teng Joon Lim
In this paper, we focus on the problem of remote state estimation in wireless networked cyber-physical systems (CPS). Information from multiple sensors is transmitted to a central gateway over a wireless network with fewer channels than sensors. Channel and power allocation are performed jointly, in the presence of a denial of service (DoS) attack where one or more channels are jammed by an attacker transmitting spurious signals. The attack policy is unknown and the central gateway has the objective of minimizing state estimation error with maximum energy efficiency. The problem involves a combination of discrete and continuous action spaces. In addition, the state and action spaces have high dimensionality, and the channel states are not fully known to the defender. We propose an innovative model-free deep reinforcement learning (DRL) algorithm to address the problem. In addition, we develop a deep learning-based method with a novel deep neural network (DNN) structure for detecting changes in the attack policy post-training. The proposed online policy change detector accelerates the adaptation of the defender to a new attack policy and also saves computational resources compared to continuous training. In short, a complete system featuring a DRL-based defender that is trained initially and adapts continually to changes in attack policy has been developed. Our numerical results show that the proposed intelligent system can significantly enhance the resilience of the system to DoS attacks.
本文重点讨论无线网络网络物理系统（CPS）中的远程状态估计问题。来自多个传感器的信息通过无线网络传输到中央网关，其信道数量少于传感器数量。在受到拒绝服务（DoS）攻击时，一个或多个信道会受到攻击者发射的虚假信号的干扰，信道和功率分配将联合执行。攻击策略是未知的，中央网关的目标是以最大的能效最小化状态估计误差。该问题涉及离散和连续行动空间的组合。此外，状态和行动空间的维度都很高，而且防御者并不完全了解信道状态。我们提出了一种创新的无模型深度强化学习（DRL）算法来解决这个问题。此外，我们还开发了一种基于深度学习的方法，该方法采用新型深度神经网络（DNN）结构，用于检测训练后攻击策略的变化。所提出的在线策略变化检测器加快了防御者对新攻击策略的适应，与持续训练相比还节省了计算资源。总之，我们开发出了一个完整的系统，它具有基于 DRL 的防御器，该防御器经过初始训练，并能不断适应攻击策略的变化。我们的数值结果表明，所提出的智能系统能显著增强系统对 DoS 攻击的抵御能力。
{"title":"Deep Learning for Radio Resource Allocation Under DoS Attack","authors":"Ke Wang;Wanchun Liu;Teng Joon Lim","doi":"10.1109/TMLCN.2024.3403513","DOIUrl":"https://doi.org/10.1109/TMLCN.2024.3403513","url":null,"abstract":"In this paper, we focus on the problem of remote state estimation in wireless networked cyber-physical systems (CPS). Information from multiple sensors is transmitted to a central gateway over a wireless network with fewer channels than sensors. Channel and power allocation are performed jointly, in the presence of a denial of service (DoS) attack where one or more channels are jammed by an attacker transmitting spurious signals. The attack policy is unknown and the central gateway has the objective of minimizing state estimation error with maximum energy efficiency. The problem involves a combination of discrete and continuous action spaces. In addition, the state and action spaces have high dimensionality, and the channel states are not fully known to the defender. We propose an innovative model-free deep reinforcement learning (DRL) algorithm to address the problem. In addition, we develop a deep learning-based method with a novel deep neural network (DNN) structure for detecting changes in the attack policy post-training. The proposed online policy change detector accelerates the adaptation of the defender to a new attack policy and also saves computational resources compared to continuous training. In short, a complete system featuring a DRL-based defender that is trained initially and adapts continually to changes in attack policy has been developed. Our numerical results show that the proposed intelligent system can significantly enhance the resilience of the system to DoS attacks.","PeriodicalId":100641,"journal":{"name":"IEEE Transactions on Machine Learning in Communications and Networking","volume":"2 ","pages":"703-716"},"PeriodicalIF":0.0,"publicationDate":"2024-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10535299","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141245171","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-16DOI: 10.1109/TMLCN.2024.3402178
Ryan M. Dreifuerst;Robert W. Heath
Beam management is a strategy to unify beamforming and channel state information (CSI) acquisition with large antenna arrays in 5G. Codebooks serve multiple uses in beam management including beamforming reference signals, CSI reporting, and analog beam training. In this paper, we propose and evaluate a machine learning-refined codebook design process for extremely large multiple-input multiple-output (X-MIMO) systems. We propose a neural network and beam selection strategy to design the initial access and refinement codebooks using end-to-end learning from beamspace representations. The algorithm, called Extreme-Beam Management ( $text {X-BM}$ ), can significantly improve the performance of extremely large arrays as envisioned for 6G and capture realistic wireless and physical layer aspects. Our results show an 8dB improvement in initial access and overall effective spectral efficiency improvements compared to traditional codebook methods.
{"title":"Hierarchical ML Codebook Design for Extreme MIMO Beam Management","authors":"Ryan M. Dreifuerst;Robert W. Heath","doi":"10.1109/TMLCN.2024.3402178","DOIUrl":"https://doi.org/10.1109/TMLCN.2024.3402178","url":null,"abstract":"Beam management is a strategy to unify beamforming and channel state information (CSI) acquisition with large antenna arrays in 5G. Codebooks serve multiple uses in beam management including beamforming reference signals, CSI reporting, and analog beam training. In this paper, we propose and evaluate a machine learning-refined codebook design process for extremely large multiple-input multiple-output (X-MIMO) systems. We propose a neural network and beam selection strategy to design the initial access and refinement codebooks using end-to-end learning from beamspace representations. The algorithm, called Extreme-Beam Management (\u0000<inline-formula> <tex-math>$text {X-BM}$ </tex-math></inline-formula>\u0000), can significantly improve the performance of extremely large arrays as envisioned for 6G and capture realistic wireless and physical layer aspects. Our results show an 8dB improvement in initial access and overall effective spectral efficiency improvements compared to traditional codebook methods.","PeriodicalId":100641,"journal":{"name":"IEEE Transactions on Machine Learning in Communications and Networking","volume":"2 ","pages":"688-702"},"PeriodicalIF":0.0,"publicationDate":"2024-03-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10533211","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141245179","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-16DOI: 10.1109/TMLCN.2024.3400756
Abdulmomen Ghalkha;Chaouki Ben Issaid;Anis Elgabli;Mehdi Bennis
This paper tackles a challenging decentralized consensus optimization problem defined over a network of interconnected devices. The devices work collaboratively to solve a problem using only their local data and exchanging information with their immediate neighbors. One approach to solving such a problem is to use Newton-type methods, which are known for their fast convergence. However, these methods have a significant drawback as they require transmitting Hessian information between devices. This not only makes them communication-inefficient but also raises privacy concerns. To address these issues, we present a novel approach that transforms the Newton direction learning problem into a formulation composed of a sum of separable functions subjected to a consensus constraint and learns an inexact Newton direction alongside the global model without enforcing devices to share their computed Hessians using the proximal primal-dual (Prox-PDA) algorithm. Our algorithm, coined DIN, avoids sharing Hessian information between devices since each device shares a model-sized vector, concealing the first- and second-order information, reducing the network’s burden and improving both communication and energy efficiencies. Furthermore, we prove that DIN descent direction converges linearly to the optimal Newton direction. Numerical simulations corroborate that DIN exhibits higher communication efficiency in terms of communication rounds while consuming less communication and computation energy compared to existing second-order decentralized baselines.
本文探讨了一个具有挑战性的分散式共识优化问题，该问题是在一个由相互连接的设备组成的网络上定义的。这些设备协同工作，仅使用其本地数据并与其近邻交换信息来解决问题。解决此类问题的一种方法是使用牛顿式方法，这种方法以收敛速度快而著称。然而，这些方法有一个明显的缺点，即它们需要在设备之间传输 Hessian 信息。这不仅降低了通信效率，还引发了隐私问题。为了解决这些问题，我们提出了一种新方法，它将牛顿方向学习问题转化为一个由受共识约束的可分离函数之和组成的表述，并使用近似基元-二元（Prox-PDA）算法，在全局模型旁学习一个不精确的牛顿方向，而不强制设备共享其计算的 Hessians。我们的算法被称为 DIN 算法，由于每个设备共享一个模型大小的向量，因此避免了设备间共享 Hessian 信息，从而隐藏了一阶和二阶信息，减轻了网络负担，提高了通信和能效。此外，我们还证明 DIN 下降方向线性收敛于最优牛顿方向。数值模拟证实，与现有的二阶分散基线相比，DIN 在通信轮数方面表现出更高的通信效率，同时消耗更少的通信和计算能量。
{"title":"DIN: A Decentralized Inexact Newton Algorithm for Consensus Optimization","authors":"Abdulmomen Ghalkha;Chaouki Ben Issaid;Anis Elgabli;Mehdi Bennis","doi":"10.1109/TMLCN.2024.3400756","DOIUrl":"https://doi.org/10.1109/TMLCN.2024.3400756","url":null,"abstract":"This paper tackles a challenging decentralized consensus optimization problem defined over a network of interconnected devices. The devices work collaboratively to solve a problem using only their local data and exchanging information with their immediate neighbors. One approach to solving such a problem is to use Newton-type methods, which are known for their fast convergence. However, these methods have a significant drawback as they require transmitting Hessian information between devices. This not only makes them communication-inefficient but also raises privacy concerns. To address these issues, we present a novel approach that transforms the Newton direction learning problem into a formulation composed of a sum of separable functions subjected to a consensus constraint and learns an inexact Newton direction alongside the global model without enforcing devices to share their computed Hessians using the proximal primal-dual (Prox-PDA) algorithm. Our algorithm, coined DIN, avoids sharing Hessian information between devices since each device shares a model-sized vector, concealing the first- and second-order information, reducing the network’s burden and improving both communication and energy efficiencies. Furthermore, we prove that DIN descent direction converges linearly to the optimal Newton direction. Numerical simulations corroborate that DIN exhibits higher communication efficiency in terms of communication rounds while consuming less communication and computation energy compared to existing second-order decentralized baselines.","PeriodicalId":100641,"journal":{"name":"IEEE Transactions on Machine Learning in Communications and Networking","volume":"2 ","pages":"663-674"},"PeriodicalIF":0.0,"publicationDate":"2024-03-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10531222","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141181872","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
During the past 10 years, researchers have extensively explored the use of machine learning (ML) in enhancing network intrusion detection systems (IDS). While many studies focused on improving accuracy of ML-based IDS, true effectiveness lies in robust generalization: the ability to classify unseen data accurately. Many existing models train and test on the same dataset, failing to represent the real unseen scenarios. Others who train and test using different datasets often struggle to generalize effectively. This study emphasizes the improvement of generalization through a novel composite approach involving the use of a lifecycle-based dataset (characterizing the attack as sequences of techniques), automatic feature learning (auto-learning), and a CNN-based deep learning model. The established model is tested on five public datasets to assess its generalization performance. The proposed approach demonstrates outstanding generalization performance, achieving an average F1 score of 0.85 and a recall of 0.94. This significantly outperforms the 0.56 and 0.42 averages recall achieved by attack-based datasets using CIC-IDS-2017 and CIC-IDS-2018 as training data, respectively. Furthermore, auto-learning features boost the F1 score by 0.2 compared to traditional statistical features. Overall, the efforts have resulted in significant advancements in model generalization, offering a more robust strategy for addressing intrusion detection challenges.
在过去 10 年中，研究人员广泛探索了如何利用机器学习（ML）来增强网络入侵检测系统（IDS）。虽然许多研究侧重于提高基于 ML 的 IDS 的准确性，但真正的有效性在于强大的泛化能力：对未见数据进行准确分类的能力。许多现有模型在相同的数据集上进行训练和测试，无法代表真实的未知场景。其他使用不同数据集进行训练和测试的模型往往难以有效地泛化。本研究强调通过一种新颖的复合方法来提高泛化能力，这种方法涉及使用基于生命周期的数据集（将攻击表征为技术序列）、自动特征学习（自动学习）和基于 CNN 的深度学习模型。已建立的模型在五个公共数据集上进行了测试，以评估其泛化性能。所提出的方法展示了出色的泛化性能，平均 F1 得分为 0.85，召回率为 0.94。这明显优于使用 CIC-IDS-2017 和 CIC-IDS-2018 作为训练数据的基于攻击的数据集分别取得的 0.56 和 0.42 的平均召回率。此外，与传统统计特征相比，自动学习特征将 F1 分数提高了 0.2。总体而言，这些努力在模型泛化方面取得了显著进步，为应对入侵检测挑战提供了更稳健的策略。
{"title":"Improving Generalization of ML-Based IDS With Lifecycle-Based Dataset, Auto-Learning Features, and Deep Learning","authors":"Didik Sudyana;Ying-Dar Lin;Miel Verkerken;Ren-Hung Hwang;Yuan-Cheng Lai;Laurens D’Hooge;Tim Wauters;Bruno Volckaert;Filip De Turck","doi":"10.1109/TMLCN.2024.3402158","DOIUrl":"https://doi.org/10.1109/TMLCN.2024.3402158","url":null,"abstract":"During the past 10 years, researchers have extensively explored the use of machine learning (ML) in enhancing network intrusion detection systems (IDS). While many studies focused on improving accuracy of ML-based IDS, true effectiveness lies in robust generalization: the ability to classify unseen data accurately. Many existing models train and test on the same dataset, failing to represent the real unseen scenarios. Others who train and test using different datasets often struggle to generalize effectively. This study emphasizes the improvement of generalization through a novel composite approach involving the use of a lifecycle-based dataset (characterizing the attack as sequences of techniques), automatic feature learning (auto-learning), and a CNN-based deep learning model. The established model is tested on five public datasets to assess its generalization performance. The proposed approach demonstrates outstanding generalization performance, achieving an average F1 score of 0.85 and a recall of 0.94. This significantly outperforms the 0.56 and 0.42 averages recall achieved by attack-based datasets using CIC-IDS-2017 and CIC-IDS-2018 as training data, respectively. Furthermore, auto-learning features boost the F1 score by 0.2 compared to traditional statistical features. Overall, the efforts have resulted in significant advancements in model generalization, offering a more robust strategy for addressing intrusion detection challenges.","PeriodicalId":100641,"journal":{"name":"IEEE Transactions on Machine Learning in Communications and Networking","volume":"2 ","pages":"645-662"},"PeriodicalIF":0.0,"publicationDate":"2024-03-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10531223","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141084910","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-13DOI: 10.1109/TMLCN.2024.3377174
Stefan Baumgartner;Oliver Lang;Mario Huemer
In recent years data-driven machine learning approaches have been extensively studied to replace or enhance traditionally model-based processing in digital communication systems. In this work, we focus on equalization and propose a novel neural network (NN-)based approach, referred to as SICNN. SICNN is designed by deep unfolding a model-based iterative soft interference cancellation (SIC) method. It eliminates the main disadvantages of its model-based counterpart, which suffers from high computational complexity and performance degradation due to required approximations. We present different variants of SICNN. SICNNv1 is specifically tailored to single carrier frequency domain equalization (SC-FDE) systems, the communication system mainly regarded in this work. SICNNv2 is more universal and is applicable as an equalizer in any communication system with a block-based data transmission scheme. Moreover, for both SICNNv1 and SICNNv2, we present versions with highly reduced numbers of learnable parameters. Another contribution of this work is a novel approach for generating training datasets for NN-based equalizers, which significantly improves their performance at high signal-to-noise ratios. We compare the bit error ratio performance of the proposed NN-based equalizers with state-of-the-art model-based and NN-based approaches, highlighting the superiority of SICNNv1 over all other methods for SC-FDE. Exemplarily, to emphasize its universality, SICNNv2 is additionally applied to a unique word orthogonal frequency division multiplexing (UW-OFDM) system, where it achieves state-of-the-art performance. Furthermore, we present a thorough complexity analysis of the proposed NN-based equalization approaches, and we investigate the influence of the training set size on the performance of NN-based equalizers.
近年来，人们对数据驱动的机器学习方法进行了广泛研究，以取代或增强数字通信系统中传统的基于模型的处理方法。在这项工作中，我们重点关注均衡问题，并提出了一种基于神经网络（NN）的新方法，即 SICNN。SICNN 是通过深度展开基于模型的迭代软干扰消除（SIC）方法而设计的。它消除了基于模型的对应方法的主要缺点，即计算复杂度高和由于需要近似而导致性能下降。我们介绍了 SICNN 的不同变体。SICNNv1 专门针对单载波频域均衡（SC-FDE）系统（本研究主要考虑的通信系统）。SICNNv2 则更具通用性，可作为均衡器应用于任何采用基于块的数据传输方案的通信系统。此外，对于 SICNNv1 和 SICNNv2，我们都提出了可学习参数数量大大减少的版本。这项工作的另一个贡献是为基于 NN 的均衡器生成训练数据集的新方法，它显著提高了均衡器在高信噪比下的性能。我们将所提出的基于 NN 的均衡器的误码率性能与最先进的基于模型和基于 NN 的方法进行了比较，突出了 SICNNv1 在 SC-FDE 方面优于所有其他方法。此外，为了强调其通用性，我们还将 SICNNv2 应用于独特的字正交频分复用（UW-OFDM）系统，并在该系统中实现了最先进的性能。此外，我们还对所提出的基于 NN 的均衡方法进行了全面的复杂性分析，并研究了训练集大小对基于 NN 的均衡器性能的影响。
{"title":"SICNN: Soft Interference Cancellation Inspired Neural Network Equalizers","authors":"Stefan Baumgartner;Oliver Lang;Mario Huemer","doi":"10.1109/TMLCN.2024.3377174","DOIUrl":"https://doi.org/10.1109/TMLCN.2024.3377174","url":null,"abstract":"In recent years data-driven machine learning approaches have been extensively studied to replace or enhance traditionally model-based processing in digital communication systems. In this work, we focus on equalization and propose a novel neural network (NN-)based approach, referred to as SICNN. SICNN is designed by deep unfolding a model-based iterative soft interference cancellation (SIC) method. It eliminates the main disadvantages of its model-based counterpart, which suffers from high computational complexity and performance degradation due to required approximations. We present different variants of SICNN. SICNNv1 is specifically tailored to single carrier frequency domain equalization (SC-FDE) systems, the communication system mainly regarded in this work. SICNNv2 is more universal and is applicable as an equalizer in any communication system with a block-based data transmission scheme. Moreover, for both SICNNv1 and SICNNv2, we present versions with highly reduced numbers of learnable parameters. Another contribution of this work is a novel approach for generating training datasets for NN-based equalizers, which significantly improves their performance at high signal-to-noise ratios. We compare the bit error ratio performance of the proposed NN-based equalizers with state-of-the-art model-based and NN-based approaches, highlighting the superiority of SICNNv1 over all other methods for SC-FDE. Exemplarily, to emphasize its universality, SICNNv2 is additionally applied to a unique word orthogonal frequency division multiplexing (UW-OFDM) system, where it achieves state-of-the-art performance. Furthermore, we present a thorough complexity analysis of the proposed NN-based equalization approaches, and we investigate the influence of the training set size on the performance of NN-based equalizers.","PeriodicalId":100641,"journal":{"name":"IEEE Transactions on Machine Learning in Communications and Networking","volume":"2 ","pages":"384-406"},"PeriodicalIF":0.0,"publicationDate":"2024-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10471626","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140297097","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-13DOI: 10.1109/TMLCN.2024.3400059
Jiaxin Li;Peng Yi;Tong Duan;Zhen Zhang;Tao Hu
Collective motion is an important research content in the multi-agent control field. However, existing multi-agent collective motion methods typically assume large communication ranges of individual agents; in the scenario of leader-follower control with short communication ranges, if the leader dynamically changes its velocity without considering the followers’ states, the communication topology may be easily disconnected, making multi-agent collective motion more challenging. In this work, a novel Hierarchical DeepDeterministic PolicyGradient (HDDPG) based reinforcement learning framework is proposed to realize multi-agent collective motion with short communication ranges, ensuring the communication topology connected as much as possible. In H-DDPG, multiple agents with one single leader and numerous followers are dynamically divided into several hierarchies to conduct distributed control when the leader’s velocity changes. Two algorithms based on DDPG and the hierarchical strategy are designed to train followers in the first layer and followers in layers other than the first layer separately, which ensures that the agents form a tight swarm from scattered distribution and all followers can track the leader effectively. The experimental results demonstrate that with short communication ranges, H-DDPG outperforms the hierarchical flocking method in keeping the communication topology connection and shaping a tighter swarm.
{"title":"Hierarchical DDPG Based Reinforcement Learning Framework for Multi-Agent Collective Motion With Short Communication Ranges","authors":"Jiaxin Li;Peng Yi;Tong Duan;Zhen Zhang;Tao Hu","doi":"10.1109/TMLCN.2024.3400059","DOIUrl":"https://doi.org/10.1109/TMLCN.2024.3400059","url":null,"abstract":"Collective motion is an important research content in the multi-agent control field. However, existing multi-agent collective motion methods typically assume large communication ranges of individual agents; in the scenario of leader-follower control with short communication ranges, if the leader dynamically changes its velocity without considering the followers’ states, the communication topology may be easily disconnected, making multi-agent collective motion more challenging. In this work, a novel Hierarchical DeepDeterministic PolicyGradient (HDDPG) based reinforcement learning framework is proposed to realize multi-agent collective motion with short communication ranges, ensuring the communication topology connected as much as possible. In H-DDPG, multiple agents with one single leader and numerous followers are dynamically divided into several hierarchies to conduct distributed control when the leader’s velocity changes. Two algorithms based on DDPG and the hierarchical strategy are designed to train followers in the first layer and followers in layers other than the first layer separately, which ensures that the agents form a tight swarm from scattered distribution and all followers can track the leader effectively. The experimental results demonstrate that with short communication ranges, H-DDPG outperforms the hierarchical flocking method in keeping the communication topology connection and shaping a tighter swarm.","PeriodicalId":100641,"journal":{"name":"IEEE Transactions on Machine Learning in Communications and Networking","volume":"2 ","pages":"633-644"},"PeriodicalIF":0.0,"publicationDate":"2024-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10529296","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141068966","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Several recent research papers in the Internet of Medical Things (IoMT) domain employ machine learning techniques to detect data patterns and trends, identify anomalies, predict and prevent adverse events, and develop personalized patient treatment plans. Despite the potential of machine learning techniques in IoMT to revolutionize healthcare, several challenges remain.The conventional machine learning models in the IoMT domain are static in that they were trained on some datasets and are being used for real-time inferencing data. This approach does not consider the patient’s recent health-related data. In the conventional machine learning models paradigm, the models must be re-trained again, even to incorporate a few sets of additional samples. Also, since the training of the conventional machine learning models generally happens on cloud platforms, there are also risks to security and privacy. Addressing these several issues, we propose an edge-based incremental learning framework with a novel feature selection algorithm for intelligent diagnosis of patients. The approach aims to improve the accuracy and efficiency of medical diagnosis by continuously learning from new patient data and adapting to patient conditions over time, along with reducing privacy and security issues. Addressing the issue of excessive features, which might increase the computational burden on incremental models, we propose a novel feature selection algorithm based on bijective soft sets, Shannon entropy, and TOPSIS(Technique for Order Preference by Similarity to Ideal Solution). We propose two incremental algorithms inspired by Aggregated Mondrian Forests and Half-Space Trees for classification and anomaly detection. The proposed model for classification gives an accuracy of 87.63%, which is better by 13.61% than the best-performing batch learning-based model. Similarly, the proposed model for anomaly detection gives an accuracy of 97.22%, which is better by 1.76% than the best-performing batch-based model. The proposed incremental algorithms for classification and anomaly detection are 9X and 16X faster than their corresponding best-performing batch learning-based models.
{"title":"An IoMT-Based Incremental Learning Framework With a Novel Feature Selection Algorithm for Intelligent Diagnosis in Smart Healthcare","authors":"Siva Sai;Kartikey Singh Bhandari;Aditya Nawal;Vinay Chamola;Biplab Sikdar","doi":"10.1109/TMLCN.2024.3374253","DOIUrl":"https://doi.org/10.1109/TMLCN.2024.3374253","url":null,"abstract":"Several recent research papers in the Internet of Medical Things (IoMT) domain employ machine learning techniques to detect data patterns and trends, identify anomalies, predict and prevent adverse events, and develop personalized patient treatment plans. Despite the potential of machine learning techniques in IoMT to revolutionize healthcare, several challenges remain.The conventional machine learning models in the IoMT domain are static in that they were trained on some datasets and are being used for real-time inferencing data. This approach does not consider the patient’s recent health-related data. In the conventional machine learning models paradigm, the models must be re-trained again, even to incorporate a few sets of additional samples. Also, since the training of the conventional machine learning models generally happens on cloud platforms, there are also risks to security and privacy. Addressing these several issues, we propose an edge-based incremental learning framework with a novel feature selection algorithm for intelligent diagnosis of patients. The approach aims to improve the accuracy and efficiency of medical diagnosis by continuously learning from new patient data and adapting to patient conditions over time, along with reducing privacy and security issues. Addressing the issue of excessive features, which might increase the computational burden on incremental models, we propose a novel feature selection algorithm based on bijective soft sets, Shannon entropy, and TOPSIS(Technique for Order Preference by Similarity to Ideal Solution). We propose two incremental algorithms inspired by Aggregated Mondrian Forests and Half-Space Trees for classification and anomaly detection. The proposed model for classification gives an accuracy of 87.63%, which is better by 13.61% than the best-performing batch learning-based model. Similarly, the proposed model for anomaly detection gives an accuracy of 97.22%, which is better by 1.76% than the best-performing batch-based model. The proposed incremental algorithms for classification and anomaly detection are 9X and 16X faster than their corresponding best-performing batch learning-based models.","PeriodicalId":100641,"journal":{"name":"IEEE Transactions on Machine Learning in Communications and Networking","volume":"2 ","pages":"370-383"},"PeriodicalIF":0.0,"publicationDate":"2024-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10461070","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140164171","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-02DOI: 10.1109/TMLCN.2024.3396438
Ahmad Gendia;Osamu Muta;Sherief Hashima;Kohei Hatano
This paper proposes two energy-efficient reinforcement learning (RL)-based algorithms for millimeter wave (mmWave)-enabled unmanned aerial vehicle (UAV) communications toward beyond-5G (B5G). This can be especially useful in ad-hoc communication scenarios within a neighborhood with main-network connectivity problems such as in areas affected by natural disasters. To improve the system’s overall sum-rate performance, the UAV-operated mobile base station (UAV-MBS) can harness non-orthogonal multiple access (NOMA) as an efficient protocol to grant ground devices access to fast downlink connections. Dynamic selection of suitable hovering spots within the target zone where the battery-constrained UAV needs to be positioned as well as calibrated NOMA power control with proper device pairing are critical for optimized performance. We propose cost-subsidized multiarmed bandit (CS-MAB) and double deep Q-network (DDQN)-based solutions to jointly address the problems of dynamic UAV path design, device pairing, and power splitting for downlink data transmission in NOMA-based systems. To verify that the proposed RL-based solutions support high sum-rates, numerical simulations are presented. In addition, exhaustive and random search benchmarks are provided as baselines for the achievable upper and lower sum-rate levels, respectively. The proposed DDQN agent achieves 96% of the sum-rate provided by the optimal exhaustive scanning whereas CS-MAB reaches 91.5%. By contrast, a conventional channel state sorting pairing (CSSP) solver achieves about 89.3%.
{"title":"Energy-Efficient Trajectory Planning With Joint Device Selection and Power Splitting for mmWaves-Enabled UAV-NOMA Networks","authors":"Ahmad Gendia;Osamu Muta;Sherief Hashima;Kohei Hatano","doi":"10.1109/TMLCN.2024.3396438","DOIUrl":"https://doi.org/10.1109/TMLCN.2024.3396438","url":null,"abstract":"This paper proposes two energy-efficient reinforcement learning (RL)-based algorithms for millimeter wave (mmWave)-enabled unmanned aerial vehicle (UAV) communications toward beyond-5G (B5G). This can be especially useful in ad-hoc communication scenarios within a neighborhood with main-network connectivity problems such as in areas affected by natural disasters. To improve the system’s overall sum-rate performance, the UAV-operated mobile base station (UAV-MBS) can harness non-orthogonal multiple access (NOMA) as an efficient protocol to grant ground devices access to fast downlink connections. Dynamic selection of suitable hovering spots within the target zone where the battery-constrained UAV needs to be positioned as well as calibrated NOMA power control with proper device pairing are critical for optimized performance. We propose cost-subsidized multiarmed bandit (CS-MAB) and double deep Q-network (DDQN)-based solutions to jointly address the problems of dynamic UAV path design, device pairing, and power splitting for downlink data transmission in NOMA-based systems. To verify that the proposed RL-based solutions support high sum-rates, numerical simulations are presented. In addition, exhaustive and random search benchmarks are provided as baselines for the achievable upper and lower sum-rate levels, respectively. The proposed DDQN agent achieves 96% of the sum-rate provided by the optimal exhaustive scanning whereas CS-MAB reaches 91.5%. By contrast, a conventional channel state sorting pairing (CSSP) solver achieves about 89.3%.","PeriodicalId":100641,"journal":{"name":"IEEE Transactions on Machine Learning in Communications and Networking","volume":"2 ","pages":"617-632"},"PeriodicalIF":0.0,"publicationDate":"2024-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10517756","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141068938","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-01DOI: 10.1109/TMLCN.2024.3352541
Darijo Raca;Ahmed H. Zahran;Cormac J. Sreenan;Rakesh K. Sinha;Emir Halepovic;Vijay Gopalakrishnan
AI-driven data analysis methods have garnered attention in enhancing the performance of wireless networks. One such application is the prediction of downlink throughput in mobile cellular networks. Accurate throughput predictions have demonstrated significant application benefits, such as improving the quality of experience in adaptive video streaming. However, the high degree of variability in cellular link behaviour, coupled with device mobility and diverse traffic demands, presents a complex problem. Numerous published studies have explored the application of machine learning to address this problem, displaying potential when trained and evaluated with traffic traces collected from operational networks. The focus of this paper is an empirical investigation of machine learning-based throughput prediction that runs in real-time on a smartphone, and its evaluation with video streaming in a range of real-world cellular network settings. We report on a number of key challenges that arise when performing prediction “in the wild”, dealing with practical issues one encounters with online data (not traces) and the limitations of real smartphones. These include data sampling, distribution shift, and data labelling. We describe our current solutions to these issues and quantify their efficacy, drawing lessons that we believe will be valuable to network practitioners planning to use such methodologies in operational cellular networks.
{"title":"Device-Based Cellular Throughput Prediction for Video Streaming: Lessons From a Real-World Evaluation","authors":"Darijo Raca;Ahmed H. Zahran;Cormac J. Sreenan;Rakesh K. Sinha;Emir Halepovic;Vijay Gopalakrishnan","doi":"10.1109/TMLCN.2024.3352541","DOIUrl":"https://doi.org/10.1109/TMLCN.2024.3352541","url":null,"abstract":"AI-driven data analysis methods have garnered attention in enhancing the performance of wireless networks. One such application is the prediction of downlink throughput in mobile cellular networks. Accurate throughput predictions have demonstrated significant application benefits, such as improving the quality of experience in adaptive video streaming. However, the high degree of variability in cellular link behaviour, coupled with device mobility and diverse traffic demands, presents a complex problem. Numerous published studies have explored the application of machine learning to address this problem, displaying potential when trained and evaluated with traffic traces collected from operational networks. The focus of this paper is an empirical investigation of machine learning-based throughput prediction that runs in real-time on a smartphone, and its evaluation with video streaming in a range of real-world cellular network settings. We report on a number of key challenges that arise when performing prediction “in the wild”, dealing with practical issues one encounters with online data (not traces) and the limitations of real smartphones. These include data sampling, distribution shift, and data labelling. We describe our current solutions to these issues and quantify their efficacy, drawing lessons that we believe will be valuable to network practitioners planning to use such methodologies in operational cellular networks.","PeriodicalId":100641,"journal":{"name":"IEEE Transactions on Machine Learning in Communications and Networking","volume":"2 ","pages":"318-334"},"PeriodicalIF":0.0,"publicationDate":"2024-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10457536","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140063514","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","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：