No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
mmWave Beam Selection in Analog Beamforming Using Personalized Federated Learning
... In particular, RadioML offers labeled in-phase/quadrature (IQ) signal recordings, which are utilized by the MoE-based automatic modulation classification (AMC) model to capture distinctive signal features under varying SINR scenarios [244]. Meanwhile, DeepMIMO, built on accurate ray-tracing simulations, provides realistic mmWave and massive MIMO channels tailored to beam selection, channel estimation, and large-antenna array optimizations [245]. For network security, 5G-NIDD presents comprehensive network intrusion detection logs with fully labeled traffic traces derived from operational realworld 5G testbeds. ...
Mixture of Experts (MoE) has emerged as a promising paradigm for scaling model capacity while preserving computational efficiency, particularly in large-scale machine learning architectures such as large language models (LLMs). Recent advances in MoE have facilitated its adoption in wireless networks to address the increasing complexity and heterogeneity of modern communication systems. This paper presents a comprehensive survey of the MoE framework in wireless networks, highlighting its potential in optimizing resource efficiency, improving scalability, and enhancing adaptability across diverse network tasks. We first introduce the fundamental concepts of MoE, including various gating mechanisms and the integration with generative AI (GenAI) and reinforcement learning (RL). Subsequently, we discuss the extensive applications of MoE across critical wireless communication scenarios, such as vehicular networks, unmanned aerial vehicles (UAVs), satellite communications, heterogeneous networks, integrated sensing and communication (ISAC), and mobile edge networks. Furthermore, key applications in channel prediction, physical layer signal processing, radio resource management, network optimization, and security are thoroughly examined. Additionally, we present a detailed overview of open-source datasets that are widely used in MoE-based models to support diverse machine learning tasks. Finally, this survey identifies crucial future research directions for MoE, emphasizing the importance of advanced training techniques, resource-aware gating strategies, and deeper integration with emerging 6G technologies.
Seamless and high-rate millimeter wave (mmWave) communications hinge on precise alignment between transmit and receive beams. While traditional methods like beam sweeping incur high pilot overhead, beam tracking is sensitive to movement patterns. In this study, we leverage received pilot signals collected over a time window and treat them as a time series to predict optimal beams of mmWave users. Inspired by the feature extraction capabilities of machine learning (ML) models, we propose a deep neural network (DNN) trained on a sequence of received pilot signals. Since each user’s dataset reflects only their movement pattern, we adopt federated learning (FL). Here, users train local models and transmit trained parameters to the base station (BS) for aggregation, ensuring model versatility. We assess the performance of our FL-aided beam prediction model using a popular mmWave channel dataset and study the impact of numerous key parameters. We also compare our method against state-of-the-art beam alignment approaches.
Millimeter wave massive multi input multi output (mmWave m-MIMO) holds immense potential for beyond 5G networks, however its practical implementation faces hurdles in beamforming, channel estimation, and signal detection. Deep learning (DL) has emerged as a promising solution for tackling these challenges, but centralized learning (CL) approaches raise concerns about privacy and communication overhead. This is where federated learning (FL) steps in as a compelling alternative. By distributing the workload to user devices while respecting data privacy, FL offers a promising avenue for overcoming the limitations of CL in mmWave m-MIMO applications. This paper delves into the current and future advancements of FL in the context of mmWave m-MIMO. We specifically focus on how FL can be applied to channel estimation, beamforming, and signal detection within this framework. Furthermore, we summarize relevant open source dataset, code, and FL framework to facilitate the development of FL for tackling challenges in mmWave m-MIMO. To conclude, we analyze the key research challenges and opportunities that lie ahead for the development of FL in mmWave m-MIMO applications.
Benefiting from huge bandwidth resources, millimeter-wave (mmWave) communications provide one of the most promising technologies for next-generation wireless networks. To compensate for the high pathloss of mmWave
signals, large-scale antenna arrays are required both at the base stations and user equipment to establish directional beamforming, where beam-management is adopted to acquire and track the optimal beam pair having the maximum received power. Naturally, narrow beams are required for achieving high beamforming gain, but they impose enormous training overhead and high sensitivity to blockages. As a remedy, deep learning (DL) may be harnessed for beam-management. First, the current state-of-the-art is reviewed, followed by the associated challenges and future research opportunities. We conclude by highlighting the associated DL design insights and novel beam-management mechanisms.
Accurate beamforming is a critical challenge for mmWave communications. Because of the large training overhead of beam training at high frequencies, it becomes relevant to exploit the available knowledge at sub-6GHz to predict the mmWave beamforming vectors using deep learning tools. In addition, fully centralized learning (CL) approaches require training over all the users data, rising major issues in terms of signaling and computational cost. To address these issues, we propose a federated learning (FL) scheme in a wireless network composed of multiple communicating links (access points-users) to predict directly the downlink mmWave beamforming vectors from the uplink sub-6GHz channels. The access points train their local deep neural networks using local data and only share their model parameters to obtain an average global one, which improves the quality of their prediction in terms of data rate. Our experiments demonstrate the potential and robustness of our proposed scheme especially under difficult conditions, performing close to the fully centralized one. When the training data is scarce, the relative gain of our scheme can reach up to 50% compared to a fully distributed one. Remarkably, our scheme can even outperform the fully centralized one when the quality of the training data is poor, enjoying a relative gain of up to 14%.
Machine learning for hybrid beamforming has been extensively studied by using centralized machine learning (CML) techniques, which require the training of a global model with a large dataset collected from the users. However, the transmission of the whole dataset between the users and the base station (BS) is computationally prohibitive due to limited communication bandwidth and privacy concerns. In this work, we introduce a federated learning (FL) based framework for hybrid beamforming, where the model training is performed at the BS by collecting only the gradients from the users. We design a convolutional neural network, in which the input is the channel data, yielding the analog beamformers at the output. Via numerical simulations, FL is demonstrated to be more tolerant to the imperfections and corruptions in the channel data as well as having less transmission overhead than CML.
Federated Learning (FL) is a promising framework for distributed learning when data is private and sensitive. However, the state-of-the-art solutions in this framework are not optimal when data is heterogeneous and non-IID. We propose a practical and robust approach to personalization in FL that adjusts to heterogeneous and non-IID data by balancing exploration and exploitation of several global models. To achieve our aim of personalization, we use a Mixture of Experts (MoE) that learns to group clients that are similar to each other, while using the global models more efficiently. We show that our approach achieves an accuracy up to 29.78% better than the state-of-the-art and up to 4.38% better compared to a local model in a pathological non-IID setting, even though we tune our approach in the IID setting.KeywordsFederated learningPersonalizationPrivacy preserving
With the increasing sensing and computing power of vehicles, beam selection solutions based on deep learning (DL) and sensor data have attracted attention in vehicle-to-infrastructure (V2I) scenarios. The existing DL-based beam selection solutions generally provide a single global model for all vehicles. However, since the data distributions across vehicles are generally different in practice, the single model may not be suitable for all vehicles. In this letter, we propose a novel personalized beam selection solution, in which each user has a tailored model. Furthermore, we propose a mask-based pre-training and fine-tuning algorithm to accomplish the personalization for beam selection. Simulation results demonstrate that the proposed personalized solution has better performance than the conventional baselines.
We address the problem of federated learning (FL) where users are distributed and partitioned into clusters. This setup captures settings where different groups of users have their own objectives (learning tasks) but by aggregating their data with others in the same cluster (same learning task), they can leverage the strength in numbers in order to perform more efficient federated learning. For this new framework of clustered federated learning, we propose the Iterative Federated Clustering Algorithm (IFCA), which alternately estimates the cluster identities of the users and optimizes model parameters for the user clusters via gradient descent. We analyze the convergence rate of this algorithm first in a linear model with squared loss and then for generic strongly convex and smooth loss functions. We show that in both settings, with good initialization, IFCA is guaranteed to converge, and discuss the optimality of the statistical error rate. In particular, for the linear model with two clusters, we can guarantee that our algorithm converges as long as the initialization is slightly better than random. When the clustering structure is ambiguous, we propose to train the models by combining IFCA with the weight sharing technique in multi-task learning. In the experiments, we show that our algorithm can succeed even if we relax the requirements on initialization with random initialization and multiple restarts. We also present experimental results showing that our algorithm is efficient in non-convex problems such as neural networks. We demonstrate the benefits of IFCA over the baselines on several clustered FL benchmarks.
Side information, like light detection and ranging data, is promising to help the millimeter wave (mmWave) system achieve efficient link configuration through machine learning methods. However, collecting and using this information may violate user privacy. In this letter, we propose a novel privacy-preserved split learning (SL) solution for the beam selection problem, in which the raw data is not uploaded during training and inference. In particular, it uses the proposed feature mix method to get better generalization performance and robustness to non-independently identically distribution (non-iid) data. Extensive experiments demonstrate that the proposed method outperforms learning-based baselines (e.g. the original SL and federated learning) in a variety of settings.
Due to large amounts of available spectrum at high frequencies, millimeter-wave (mmWave) technology has gained significant interest in 5G communications, but mmWave links suffer from sever free space attenuation. Codebook-based beamforming techniques with multiple antennas can effectively alleviate this challenge with low computational complexity and low cost hardware. However, small delay and high-speed communications with beamforming techniques require beam alignment with small overhead to establish the wireless link quickly. In this paper, we propose a deep learning-based low overhead analog beam selection scheme inspired by super-resolution. Deep neural networks are employed to conduct beam quality estimation based on partial beam measurements. Our proposed scheme can cover all the directions of arriving signals with low overhead by utilizing codebooks with different beam widths. Also, for the purpose of further reducing overhead, we build the model predicting beam qualities based on the past beam sweepings. With these beam quality estimation and prediction model, the beam that achieves large signal-to-noise-power-ratio (SNR) can be selected based on partial beam measurements. Simulation results show that the proposed scheme can accurately estimate beam qualities and give high probability of optimal beam selections with low overhead.
Federated learning (FL) is a machine learning setting where many clients (e.g., mobile devices or whole organizations) collaboratively train a model under the orchestration of a central server (e.g., service provider), while keeping the training data decentralized. FL embodies the principles of focused data collection and minimization, and can mitigate many of the systemic privacy risks and costs resulting from traditional, centralized machine learning and data science approaches. Motivated by the explosive growth in FL research, this monograph discusses recent advances and presents an extensive collection of open problems and challenges.
Predicting the millimeter wave (mmWave) beams and blockages using sub-6 GHz channels has the potential of enabling mobility and reliability in scalable mmWave systems. Prior work has focused on extracting spatial channel characteristics at the sub-6 GHz band and then use them to reduce the mmWave beam training overhead. This approach still requires beam refinement at mmWave and does not normally account for the different dielectric properties at the different bands. In this paper, we first prove that under certain conditions, there exist mapping functions that can predict the optimal mmWave beam and blockage status directly from the sub-6 GHz channel. These mapping functions, however, are hard to characterize analytically which motivates exploiting deep neural network models to learn them. For that, we prove that a large enough neural network can predict mmWave beams and blockages with success probabilities that can be made arbitrarily close to one. Then, we develop a deep learning model and empirically evaluate its beam/blockage prediction performance using a publicly available dataset. The results show that the proposed solution can predict the mmWave blockages with more than 90% success probability and can predict the optimal mmWave beams to approach the upper bounds while requiring no beam training overhead.
Towards Federated Learning at Scale: System Design
- K A Bonawitz
The Non-IID Data Quagmire of Decentralized Machine Learning
- K Hsieh
- A Phanishayee
- O Mutlu
- P B Gibbons
Communication-Efficient Learning of Deep Networks from Decentralized Data
- McMahan
DeepMIMO: A Generic Deep Learning Dataset for Millimeter Wave and Massive MIMO Applications
- A Alkhateeb
Adaptive Federated Optimization
- S J Reddi
The Non-IID Data Quagmire of Decentralized Machine Learning
- Hsieh
Communication-Efficient Learning of Deep Networks from Decentralized Data
- B Mcmahan
- E Moore
- D Ramage
- S Hampson
- B A Arcas
Personalized Federated Learning for mmWave Beam Prediction Using Non-IID Sub-6 GHz Channels
- Y Cheng
Symbolic Discovery of Optimization Algorithms
- X Chen