Pub Date : 2024-12-25DOI: 10.1109/TMLCN.2024.3522872
Mehdi Letafati;Samad Ali;Matti Latva-Aho
In this paper, conditional denoising diffusion probabilistic models (CDiffs) are proposed to enhance the data transmission and reconstruction over wireless channels. The underlying mechanism of diffusion models is to decompose the data generation process over the so-called “denoising” steps. Inspired by this, the key idea is to leverage the generative prior of diffusion models in learning a “noisy-to-clean” transformation of the information signal to help enhance data reconstruction. The proposed scheme could be beneficial for communication scenarios in which a prior knowledge of the information content is available, e.g., in multimedia transmission. Hence, instead of employing complicated channel codes that reduce the information rate, one can exploit diffusion priors for reliable data reconstruction, especially under extreme channel conditions due to low signal-to-noise ratio (SNR), or hardware-impaired communications. The proposed CDiff-assisted receiver is tailored for the scenario of wireless image transmission using MNIST dataset. Our numerical results highlight the reconstruction performance of our scheme compared to the conventional digital communication, as well as the deep neural network (DNN)-based benchmark. It is also shown that more than 10 dB improvement in the reconstruction could be achieved in low SNR regimes, without the need to reduce the information rate for error correction.
{"title":"Conditional Denoising Diffusion Probabilistic Models for Data Reconstruction Enhancement in Wireless Communications","authors":"Mehdi Letafati;Samad Ali;Matti Latva-Aho","doi":"10.1109/TMLCN.2024.3522872","DOIUrl":"https://doi.org/10.1109/TMLCN.2024.3522872","url":null,"abstract":"In this paper, conditional denoising diffusion probabilistic models (CDiffs) are proposed to enhance the data transmission and reconstruction over wireless channels. The underlying mechanism of diffusion models is to decompose the data generation process over the so-called “denoising” steps. Inspired by this, the key idea is to leverage the generative prior of diffusion models in learning a “noisy-to-clean” transformation of the information signal to help enhance data reconstruction. The proposed scheme could be beneficial for communication scenarios in which a prior knowledge of the information content is available, e.g., in multimedia transmission. Hence, instead of employing complicated channel codes that reduce the information rate, one can exploit diffusion priors for reliable data reconstruction, especially under extreme channel conditions due to low signal-to-noise ratio (SNR), or hardware-impaired communications. The proposed CDiff-assisted receiver is tailored for the scenario of wireless image transmission using MNIST dataset. Our numerical results highlight the reconstruction performance of our scheme compared to the conventional digital communication, as well as the deep neural network (DNN)-based benchmark. It is also shown that more than 10 dB improvement in the reconstruction could be achieved in low SNR regimes, without the need to reduce the information rate for error correction.","PeriodicalId":100641,"journal":{"name":"IEEE Transactions on Machine Learning in Communications and Networking","volume":"3 ","pages":"133-146"},"PeriodicalIF":0.0,"publicationDate":"2024-12-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10816175","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142912491","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 : 2024-12-23DOI: 10.1109/TMLCN.2024.3521876
Danish Rizvi;David Boyle
Unmanned Aerial Vehicles (UAVs) are increasingly used as aerial base stations to provide ad hoc communications infrastructure. Building upon prior research efforts which consider either static nodes, 2D trajectories or single UAV systems, this paper focuses on the use of multiple UAVs for providing wireless communication to mobile users in the absence of terrestrial communications infrastructure. In particular, we jointly optimize UAV 3D trajectory and NOMA power allocation to maximize system throughput. Firstly, a weighted K-means-based clustering algorithm establishes UAV-user associations at regular intervals. Then the efficacy of training a novel Shared Deep Q-Network (SDQN) with action masking is explored. Unlike training each UAV separately using DQN, the SDQN reduces training time by using the experiences of multiple UAVs instead of a single agent. We also show that SDQN can be used to train a multi-agent system with differing action spaces. Simulation results confirm that: 1) training a shared DQN outperforms a conventional DQN in terms of maximum system throughput (+20%) and training time (-10%); 2) it can converge for agents with different action spaces, yielding a 9% increase in throughput compared to Mutual DQN algorithm; and 3) combining NOMA with an SDQN architecture enables the network to achieve a better sum rate compared with existing baseline schemes.
{"title":"Multi-Agent Reinforcement Learning With Action Masking for UAV-Enabled Mobile Communications","authors":"Danish Rizvi;David Boyle","doi":"10.1109/TMLCN.2024.3521876","DOIUrl":"https://doi.org/10.1109/TMLCN.2024.3521876","url":null,"abstract":"Unmanned Aerial Vehicles (UAVs) are increasingly used as aerial base stations to provide ad hoc communications infrastructure. Building upon prior research efforts which consider either static nodes, 2D trajectories or single UAV systems, this paper focuses on the use of multiple UAVs for providing wireless communication to mobile users in the absence of terrestrial communications infrastructure. In particular, we jointly optimize UAV 3D trajectory and NOMA power allocation to maximize system throughput. Firstly, a weighted K-means-based clustering algorithm establishes UAV-user associations at regular intervals. Then the efficacy of training a novel Shared Deep Q-Network (SDQN) with action masking is explored. Unlike training each UAV separately using DQN, the SDQN reduces training time by using the experiences of multiple UAVs instead of a single agent. We also show that SDQN can be used to train a multi-agent system with differing action spaces. Simulation results confirm that: 1) training a shared DQN outperforms a conventional DQN in terms of maximum system throughput (+20%) and training time (-10%); 2) it can converge for agents with different action spaces, yielding a 9% increase in throughput compared to Mutual DQN algorithm; and 3) combining NOMA with an SDQN architecture enables the network to achieve a better sum rate compared with existing baseline schemes.","PeriodicalId":100641,"journal":{"name":"IEEE Transactions on Machine Learning in Communications and Networking","volume":"3 ","pages":"117-132"},"PeriodicalIF":0.0,"publicationDate":"2024-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10812765","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142905826","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 : 2024-12-16DOI: 10.1109/TMLCN.2024.3517619
Zhanwei Yu;Yi Zhao;Xiaoli Chu;Di Yuan
Passively cooled base stations (PCBSs) have emerged to deliver better cost and energy efficiency. However, passive cooling necessitates intelligent thermal control via traffic management, i.e., the instantaneous data traffic or throughput of a PCBS directly impacts its thermal performance. This is particularly challenging for outdoor deployment of PCBSs because the heat dissipation efficiency is uncertain and fluctuates over time. What is more, the PCBSs are interference-coupled in multi-cell scenarios. Thus, a higher-throughput PCBS leads to higher interference to the other PCBSs, which, in turn, would require more resource consumption to meet their respective throughput targets. In this paper, we address online decision-making for maximizing the total downlink throughput for a multi-PCBS system subject to constraints related on operating temperature. We demonstrate that a reinforcement learning (RL) approach, specifically soft actor-critic (SAC), can successfully perform throughput maximization while keeping the PCBSs cool, by adapting the throughput to time-varying heat dissipation conditions. Furthermore, we design a denial and reward mechanism that effectively mitigates the risk of overheating during the exploration phase of RL. Simulation results show that our approach achieves up to 88.6% of the global optimum. This is very promising, as our approach operates without prior knowledge of future heat dissipation efficiency, which is required by the global optimum.
{"title":"Online Learning for Intelligent Thermal Management of Interference-Coupled and Passively Cooled Base Stations","authors":"Zhanwei Yu;Yi Zhao;Xiaoli Chu;Di Yuan","doi":"10.1109/TMLCN.2024.3517619","DOIUrl":"https://doi.org/10.1109/TMLCN.2024.3517619","url":null,"abstract":"Passively cooled base stations (PCBSs) have emerged to deliver better cost and energy efficiency. However, passive cooling necessitates intelligent thermal control via traffic management, i.e., the instantaneous data traffic or throughput of a PCBS directly impacts its thermal performance. This is particularly challenging for outdoor deployment of PCBSs because the heat dissipation efficiency is uncertain and fluctuates over time. What is more, the PCBSs are interference-coupled in multi-cell scenarios. Thus, a higher-throughput PCBS leads to higher interference to the other PCBSs, which, in turn, would require more resource consumption to meet their respective throughput targets. In this paper, we address online decision-making for maximizing the total downlink throughput for a multi-PCBS system subject to constraints related on operating temperature. We demonstrate that a reinforcement learning (RL) approach, specifically soft actor-critic (SAC), can successfully perform throughput maximization while keeping the PCBSs cool, by adapting the throughput to time-varying heat dissipation conditions. Furthermore, we design a denial and reward mechanism that effectively mitigates the risk of overheating during the exploration phase of RL. Simulation results show that our approach achieves up to 88.6% of the global optimum. This is very promising, as our approach operates without prior knowledge of future heat dissipation efficiency, which is required by the global optimum.","PeriodicalId":100641,"journal":{"name":"IEEE Transactions on Machine Learning in Communications and Networking","volume":"3 ","pages":"64-79"},"PeriodicalIF":0.0,"publicationDate":"2024-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10802970","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142858862","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 : 2024-12-16DOI: 10.1109/TMLCN.2024.3517613
Aleksandar Pasquini;Rajesh Vasa;Irini Logothetis;Hassan Habibi Gharakheili;Alexander Chambers;Minh Tran
Machine Learning (ML)-based techniques are increasingly used for network management tasks, such as intrusion detection, application identification, or asset management. Recent studies show that neural network-based traffic analysis can achieve performance comparable to human feature-engineered ML pipelines. However, neural networks provide this performance at a higher computational cost and complexity, due to high-throughput traffic conditions necessitating specialized hardware for real-time operations. This paper presents lightweight models for encoding characteristics of Internet-of-Things (IoT) network packets; 1) we present two strategies to encode packets (regardless of their size, encryption, and protocol) to integer vectors: a shallow lightweight neural network and compression. With a public dataset containing about 8 million packets emitted by 22 IoT device types, we show the encoded packets can form complete (up to 80%) and homogeneous (up to 89%) clusters; 2) we demonstrate the efficacy of our generated encodings in the downstream classification task and quantify their computing costs. We train three multi-class models to predict the IoT class given network packets and show our models can achieve the same levels of accuracy (94%) as deep neural network embeddings but with computing costs up to 10 times lower; 3) we examine how the amount of packet data (headers and payload) can affect the prediction quality. We demonstrate how the choice of Internet Protocol (IP) payloads strikes a balance between prediction accuracy (99%) and cost. Along with the cost-efficacy of models, this capability can result in rapid and accurate predictions, meeting the requirements of network operators.
{"title":"Robust and Lightweight Modeling of IoT Network Behaviors From Raw Traffic Packets","authors":"Aleksandar Pasquini;Rajesh Vasa;Irini Logothetis;Hassan Habibi Gharakheili;Alexander Chambers;Minh Tran","doi":"10.1109/TMLCN.2024.3517613","DOIUrl":"https://doi.org/10.1109/TMLCN.2024.3517613","url":null,"abstract":"Machine Learning (ML)-based techniques are increasingly used for network management tasks, such as intrusion detection, application identification, or asset management. Recent studies show that neural network-based traffic analysis can achieve performance comparable to human feature-engineered ML pipelines. However, neural networks provide this performance at a higher computational cost and complexity, due to high-throughput traffic conditions necessitating specialized hardware for real-time operations. This paper presents lightweight models for encoding characteristics of Internet-of-Things (IoT) network packets; 1) we present two strategies to encode packets (regardless of their size, encryption, and protocol) to integer vectors: a shallow lightweight neural network and compression. With a public dataset containing about 8 million packets emitted by 22 IoT device types, we show the encoded packets can form complete (up to 80%) and homogeneous (up to 89%) clusters; 2) we demonstrate the efficacy of our generated encodings in the downstream classification task and quantify their computing costs. We train three multi-class models to predict the IoT class given network packets and show our models can achieve the same levels of accuracy (94%) as deep neural network embeddings but with computing costs up to 10 times lower; 3) we examine how the amount of packet data (headers and payload) can affect the prediction quality. We demonstrate how the choice of Internet Protocol (IP) payloads strikes a balance between prediction accuracy (99%) and cost. Along with the cost-efficacy of models, this capability can result in rapid and accurate predictions, meeting the requirements of network operators.","PeriodicalId":100641,"journal":{"name":"IEEE Transactions on Machine Learning in Communications and Networking","volume":"3 ","pages":"98-116"},"PeriodicalIF":0.0,"publicationDate":"2024-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10802939","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142890343","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 : 2024-12-11DOI: 10.1109/TMLCN.2024.3500756
{"title":"IEEE Communications Society Board of Governors","authors":"","doi":"10.1109/TMLCN.2024.3500756","DOIUrl":"https://doi.org/10.1109/TMLCN.2024.3500756","url":null,"abstract":"","PeriodicalId":100641,"journal":{"name":"IEEE Transactions on Machine Learning in Communications and Networking","volume":"3 ","pages":"C3-C3"},"PeriodicalIF":0.0,"publicationDate":"2024-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10792973","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142810507","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 : 2024-12-11DOI: 10.1109/TMLCN.2024.3515913
Zhizhou He;Fabien Héliot;Yi Ma
Reconfigurable Intelligent Surfaces (RIS) can enhance system performance at the cost of increased complexity in multi-user MIMO systems. The beamforming options scale with the number of antennas at the base station/RIS. Existing methods for solving this problem tend to use computationally intensive iterative methods that are non-scalable for large RIS-aided MIMO systems. We propose here a novel self-supervised contrastive learning neural network (NN) architecture to optimize the sum spectral efficiency through joint active and passive beamforming design in multi-user RIS-aided MIMO systems. Our scheme utilizes contrastive learning to capture the channel features from augmented channel data and then can be trained to perform beamforming with only 1% of labeled data. The labels are derived through a closed-form optimization algorithm, leveraging a sequential fractional programming approach. Leveraging the proposed self-supervised design helps to greatly reduce the computational complexity during the training phase. Moreover, our proposed model can operate under various noise levels by using data augmentation methods while maintaining a robust out-of-distribution performance under various propagation environments and different signal-to-noise ratios (SNR)s. During training, our proposed network only needs 10% of labeled data to converge when compared to supervised learning. Our trained NN can then achieve performance which is only $~7%$ and $~2.5%$ away from mathematical upper bound and fully supervised learning, respectively, with far less computational complexity.
{"title":"Self-Supervised Contrastive Learning for Joint Active and Passive Beamforming in RIS-Assisted MU-MIMO Systems","authors":"Zhizhou He;Fabien Héliot;Yi Ma","doi":"10.1109/TMLCN.2024.3515913","DOIUrl":"https://doi.org/10.1109/TMLCN.2024.3515913","url":null,"abstract":"Reconfigurable Intelligent Surfaces (RIS) can enhance system performance at the cost of increased complexity in multi-user MIMO systems. The beamforming options scale with the number of antennas at the base station/RIS. Existing methods for solving this problem tend to use computationally intensive iterative methods that are non-scalable for large RIS-aided MIMO systems. We propose here a novel self-supervised contrastive learning neural network (NN) architecture to optimize the sum spectral efficiency through joint active and passive beamforming design in multi-user RIS-aided MIMO systems. Our scheme utilizes contrastive learning to capture the channel features from augmented channel data and then can be trained to perform beamforming with only 1% of labeled data. The labels are derived through a closed-form optimization algorithm, leveraging a sequential fractional programming approach. Leveraging the proposed self-supervised design helps to greatly reduce the computational complexity during the training phase. Moreover, our proposed model can operate under various noise levels by using data augmentation methods while maintaining a robust out-of-distribution performance under various propagation environments and different signal-to-noise ratios (SNR)s. During training, our proposed network only needs 10% of labeled data to converge when compared to supervised learning. Our trained NN can then achieve performance which is only \u0000<inline-formula> <tex-math>$~7%$ </tex-math></inline-formula>\u0000 and \u0000<inline-formula> <tex-math>$~2.5%$ </tex-math></inline-formula>\u0000 away from mathematical upper bound and fully supervised learning, respectively, with far less computational complexity.","PeriodicalId":100641,"journal":{"name":"IEEE Transactions on Machine Learning in Communications and Networking","volume":"3 ","pages":"147-162"},"PeriodicalIF":0.0,"publicationDate":"2024-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10793234","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142912511","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 : 2024-12-09DOI: 10.1109/TMLCN.2024.3513267
Dheeraj Raja Kumar;Carles Antón-Haro;Xavier Mestre
Machine Learning tools are becoming very powerful alternatives to improve the robustness of wireless communication systems. Signal processing procedures that tend to collapse in the presence of model mismatches can be effectively improved and made robust by incorporating the selective use of data-driven techniques. This paper explores the use of neural network (NN)-based receivers to improve the reception of a Rate Splitting Multiple Access (RSMA) system. The intention is to explore several alternatives to conventional successive interference cancellation (SIC) techniques, which are known to be ineffective in the presence of channel state information (CSI) and model errors. The focus is on NN-based architectures that do not need to be retrained at each channel realization. The main idea is to replace some of the basic operations in a conventional multi-antenna SIC receiver by their NN-based equivalents, following a hybrid Model/Data-driven based approach that preserves the main procedures in the model-based signal demodulation chain. Three different architectures are explored along with their performance and computational complexity, characterized under different degrees of model uncertainty, including imperfect channel state information and non-linear channels. We evaluate the performance of data-driven architectures in overloaded scenario to analyze its effectiveness against conventional benchmarks. The study dictates that a higher degree of robustness of transceiver can be achieved, provided the neural architecture is well-designed and fed with the right information.
{"title":"Deep Receiver Architectures for Robust MIMO Rate Splitting Multiple Access","authors":"Dheeraj Raja Kumar;Carles Antón-Haro;Xavier Mestre","doi":"10.1109/TMLCN.2024.3513267","DOIUrl":"https://doi.org/10.1109/TMLCN.2024.3513267","url":null,"abstract":"Machine Learning tools are becoming very powerful alternatives to improve the robustness of wireless communication systems. Signal processing procedures that tend to collapse in the presence of model mismatches can be effectively improved and made robust by incorporating the selective use of data-driven techniques. This paper explores the use of neural network (NN)-based receivers to improve the reception of a Rate Splitting Multiple Access (RSMA) system. The intention is to explore several alternatives to conventional successive interference cancellation (SIC) techniques, which are known to be ineffective in the presence of channel state information (CSI) and model errors. The focus is on NN-based architectures that do not need to be retrained at each channel realization. The main idea is to replace some of the basic operations in a conventional multi-antenna SIC receiver by their NN-based equivalents, following a hybrid Model/Data-driven based approach that preserves the main procedures in the model-based signal demodulation chain. Three different architectures are explored along with their performance and computational complexity, characterized under different degrees of model uncertainty, including imperfect channel state information and non-linear channels. We evaluate the performance of data-driven architectures in overloaded scenario to analyze its effectiveness against conventional benchmarks. The study dictates that a higher degree of robustness of transceiver can be achieved, provided the neural architecture is well-designed and fed with the right information.","PeriodicalId":100641,"journal":{"name":"IEEE Transactions on Machine Learning in Communications and Networking","volume":"3 ","pages":"45-63"},"PeriodicalIF":0.0,"publicationDate":"2024-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10781451","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142844397","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 : 2024-12-04DOI: 10.1109/TMLCN.2024.3511475
Chenyuan Feng;Ahmed Arafa;Zihan Chen;Mingxiong Zhao;Tony Q. S. Quek;Howard H. Yang
This paper studies the efficiency of training a statistical model among an edge server and multiple clients via Federated Learning (FL) – a machine learning method that preserves data privacy in the training process – over wireless networks. Due to unreliable wireless channels and constrained communication resources, the server can only choose a handful of clients for parameter updates during each communication round. To address this issue, analytical expressions are derived to characterize the FL convergence rate, accounting for key features from both communication and algorithmic aspects, including transmission reliability, scheduling policies, and momentum method. First, the analysis reveals that either delicately designed user scheduling policies or expanding higher bandwidth to accommodate more clients in each communication round can expedite model training in networks with reliable connections. However, these methods become ineffective when the connection is erratic. Second, it has been verified that incorporating the momentum method into the model training algorithm accelerates the rate of convergence and provides greater resilience against transmission failures. Last, extensive empirical simulations are provided to verify these theoretical discoveries and enhancements in performance.
{"title":"Toward Understanding Federated Learning over Unreliable Networks","authors":"Chenyuan Feng;Ahmed Arafa;Zihan Chen;Mingxiong Zhao;Tony Q. S. Quek;Howard H. Yang","doi":"10.1109/TMLCN.2024.3511475","DOIUrl":"https://doi.org/10.1109/TMLCN.2024.3511475","url":null,"abstract":"This paper studies the efficiency of training a statistical model among an edge server and multiple clients via Federated Learning (FL) – a machine learning method that preserves data privacy in the training process – over wireless networks. Due to unreliable wireless channels and constrained communication resources, the server can only choose a handful of clients for parameter updates during each communication round. To address this issue, analytical expressions are derived to characterize the FL convergence rate, accounting for key features from both communication and algorithmic aspects, including transmission reliability, scheduling policies, and momentum method. First, the analysis reveals that either delicately designed user scheduling policies or expanding higher bandwidth to accommodate more clients in each communication round can expedite model training in networks with reliable connections. However, these methods become ineffective when the connection is erratic. Second, it has been verified that incorporating the momentum method into the model training algorithm accelerates the rate of convergence and provides greater resilience against transmission failures. Last, extensive empirical simulations are provided to verify these theoretical discoveries and enhancements in performance.","PeriodicalId":100641,"journal":{"name":"IEEE Transactions on Machine Learning in Communications and Networking","volume":"3 ","pages":"80-97"},"PeriodicalIF":0.0,"publicationDate":"2024-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10777576","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142880294","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 : 2024-11-26DOI: 10.1109/TMLCN.2024.3506874
Feng Shu;Baihua Shi;Yiwen Chen;Jiatong Bai;Yifan Li;Tingting Liu;Zhu Han;Xiaohu You
Massive multiple input multiple output (MIMO) antenna arrays eventuate a huge amount of circuit costs and computational complexity. To satisfy the needs of high precision and low cost in future green wireless communication, the conventional hybrid analog and digital MIMO receive structure emerges a natural choice. But it exists an issue of the phase ambiguity in direction of arrival (DOA) estimation and requires at least two time-slots to complete one-time DOA measurement with the first time-slot generating the set of candidate solutions and the second one to find a true direction by received beamforming over this set, which will lead to a low time-efficiency. To address this problem,a new heterogeneous sub-connected hybrid analog and digital ( $mathrm {H}^{2}$ AD) MIMO structure is proposed with an intrinsic ability of removing phase ambiguity, and then a corresponding new framework is developed to implement a rapid high-precision DOA estimation using only single time-slot. The proposed framework consists of two steps: 1) form a set of candidate solutions using existing methods like MUSIC; 2) find the class of the true solutions and compute the class mean. To infer the set of true solutions, we propose two new clustering methods: weight global minimum distance (WGMD) and weight local minimum distance (WLMD). Next, we also enhance two classic clustering methods: accelerating local weighted k-means (ALW-K-means) and improved density. Additionally, the corresponding closed-form expression of Cramer-Rao lower bound (CRLB) is derived. Simulation results show that the proposed frameworks using the above four clustering can approach the CRLB in almost all signal to noise ratio (SNR) regions except for extremely low SNR (SNR $lt -5$ dB). Four clustering methods have an accuracy decreasing order as follows: WGMD, improved DBSCAN, ALW-K-means and WLMD.
{"title":"A New Heterogeneous Hybrid Massive MIMO Receiver With an Intrinsic Ability of Removing Phase Ambiguity of DOA Estimation via Machine Learning","authors":"Feng Shu;Baihua Shi;Yiwen Chen;Jiatong Bai;Yifan Li;Tingting Liu;Zhu Han;Xiaohu You","doi":"10.1109/TMLCN.2024.3506874","DOIUrl":"https://doi.org/10.1109/TMLCN.2024.3506874","url":null,"abstract":"Massive multiple input multiple output (MIMO) antenna arrays eventuate a huge amount of circuit costs and computational complexity. To satisfy the needs of high precision and low cost in future green wireless communication, the conventional hybrid analog and digital MIMO receive structure emerges a natural choice. But it exists an issue of the phase ambiguity in direction of arrival (DOA) estimation and requires at least two time-slots to complete one-time DOA measurement with the first time-slot generating the set of candidate solutions and the second one to find a true direction by received beamforming over this set, which will lead to a low time-efficiency. To address this problem,a new heterogeneous sub-connected hybrid analog and digital (\u0000<inline-formula> <tex-math>$mathrm {H}^{2}$ </tex-math></inline-formula>\u0000AD) MIMO structure is proposed with an intrinsic ability of removing phase ambiguity, and then a corresponding new framework is developed to implement a rapid high-precision DOA estimation using only single time-slot. The proposed framework consists of two steps: 1) form a set of candidate solutions using existing methods like MUSIC; 2) find the class of the true solutions and compute the class mean. To infer the set of true solutions, we propose two new clustering methods: weight global minimum distance (WGMD) and weight local minimum distance (WLMD). Next, we also enhance two classic clustering methods: accelerating local weighted k-means (ALW-K-means) and improved density. Additionally, the corresponding closed-form expression of Cramer-Rao lower bound (CRLB) is derived. Simulation results show that the proposed frameworks using the above four clustering can approach the CRLB in almost all signal to noise ratio (SNR) regions except for extremely low SNR (SNR \u0000<inline-formula> <tex-math>$lt -5$ </tex-math></inline-formula>\u0000 dB). Four clustering methods have an accuracy decreasing order as follows: WGMD, improved DBSCAN, ALW-K-means and WLMD.","PeriodicalId":100641,"journal":{"name":"IEEE Transactions on Machine Learning in Communications and Networking","volume":"3 ","pages":"17-29"},"PeriodicalIF":0.0,"publicationDate":"2024-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10767772","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142844437","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 : 2024-11-20DOI: 10.1109/TMLCN.2024.3503543
Houssem Sifaou;Osvaldo Simeone
In many wireless application scenarios, acquiring labeled data can be prohibitively costly, requiring complex optimization processes or measurement campaigns. Semi-supervised learning leverages unlabeled samples to augment the available dataset by assigning synthetic labels obtained via machine learning (ML)-based predictions. However, treating the synthetic labels as true labels may yield worse-performing models as compared to models trained using only labeled data. Inspired by the recently developed prediction-powered inference (PPI) framework, this work investigates how to leverage the synthetic labels produced by an ML model, while accounting for the inherent bias concerning true labels. To this end, we first review PPI and its recent extensions, namely tuned PPI and cross-prediction-powered inference (CPPI). Then, we introduce two novel variants of PPI. The first, referred to as tuned CPPI, provides CPPI with an additional degree of freedom in adapting to the quality of the ML-based labels. The second, meta-CPPI (MCPPI), extends tuned CPPI via the joint optimization of the ML labeling models and of the parameters of interest. Finally, we showcase two applications of PPI-based techniques in wireless systems, namely beam alignment based on channel knowledge maps in millimeter-wave systems and received signal strength information-based indoor localization. Simulation results show the advantages of PPI-based techniques over conventional approaches that rely solely on labeled data or that apply standard pseudo-labeling strategies from semi-supervised learning. Furthermore, the proposed tuned CPPI method is observed to guarantee the best performance among all benchmark schemes, especially in the regime of limited labeled data.
{"title":"Semi-Supervised Learning via Cross-Prediction-Powered Inference for Wireless Systems","authors":"Houssem Sifaou;Osvaldo Simeone","doi":"10.1109/TMLCN.2024.3503543","DOIUrl":"https://doi.org/10.1109/TMLCN.2024.3503543","url":null,"abstract":"In many wireless application scenarios, acquiring labeled data can be prohibitively costly, requiring complex optimization processes or measurement campaigns. Semi-supervised learning leverages unlabeled samples to augment the available dataset by assigning synthetic labels obtained via machine learning (ML)-based predictions. However, treating the synthetic labels as true labels may yield worse-performing models as compared to models trained using only labeled data. Inspired by the recently developed prediction-powered inference (PPI) framework, this work investigates how to leverage the synthetic labels produced by an ML model, while accounting for the inherent bias concerning true labels. To this end, we first review PPI and its recent extensions, namely tuned PPI and cross-prediction-powered inference (CPPI). Then, we introduce two novel variants of PPI. The first, referred to as tuned CPPI, provides CPPI with an additional degree of freedom in adapting to the quality of the ML-based labels. The second, meta-CPPI (MCPPI), extends tuned CPPI via the joint optimization of the ML labeling models and of the parameters of interest. Finally, we showcase two applications of PPI-based techniques in wireless systems, namely beam alignment based on channel knowledge maps in millimeter-wave systems and received signal strength information-based indoor localization. Simulation results show the advantages of PPI-based techniques over conventional approaches that rely solely on labeled data or that apply standard pseudo-labeling strategies from semi-supervised learning. Furthermore, the proposed tuned CPPI method is observed to guarantee the best performance among all benchmark schemes, especially in the regime of limited labeled data.","PeriodicalId":100641,"journal":{"name":"IEEE Transactions on Machine Learning in Communications and Networking","volume":"3 ","pages":"30-44"},"PeriodicalIF":0.0,"publicationDate":"2024-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10758826","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142844291","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：