Comparing Stochastic and Ray-tracing Datasets in Machine Learning for Wireless Applications

Machine learning for wireless systems is commonly studied using standardized stochastic channel models (e.g., TDL/CDL/UMa) because of their legacy in wireless communication standardization and their ability to generate data at scale. However, some of their structural assumptions may diverge from real-world propagation. This paper asks when these models are sufficient and when ray-traced (RT) data - a proxy for the real world - provides tangible benefits. To answer these questions, we conduct an empirical study on two representative tasks: CSI compression and temporal channel prediction. Models are trained and evaluated using in-domain, cross-domain, and small-data fine-tuning protocols. Across settings, we observe that stochastic-only evaluation may over- or under-estimate performance relative to RT. These findings support a task-aware recipe where stochastic models can be leveraged for scalable pre-training and for tasks that do not rely on strong spatiotemporal coupling. When that coupling matters, pre-training and evaluation should be grounded in spatially consistent or geometrically similar RT scenarios. This study provides initial guidance to inform future discussions on benchmarking and standardization.

Generative Decoding of Compressed CSI for MIMO Precoding Design

Massive MIMO systems can enhance spectral and energy efficiency, but they require accurate channel state information (CSI), which becomes costly as the number of antennas increases. While machine learning (ML) autoencoders show promise for CSI reconstruction and reducing feedback overhead, they introduce new challenges with standardization, interoperability, and backward compatibility. Also, the significant data collection needed for training makes real-world deployment difficult. To overcome these drawbacks, we propose an ML-based, decoder-only solution for compressed CSI. Our approach uses a standardized encoder for CSI compression on the user side and a site-specific generative decoder at the base station to refine the compressed CSI using environmental knowledge. We introduce two training schemes for the generative decoder: An end-to-end method and a two-stage method, both utilizing a goal-oriented loss function. Furthermore, we reduce the data collection overhead by using a site-specific digital twin to generate synthetic CSI data for training. Our simulations highlight the effectiveness of this solution across various feedback overhead regimes.

Constrained Multimodal Sensing-Aided Communications: A Dynamic Beamforming Design

Using multimodal sensory data can enhance communications systems by reducing the overhead and latency in beam training. However, processing such data incurs high computational complexity, and continuous sensing results in significant power and bandwidth consumption. This gives rise to a tradeoff between the (multimodal) sensing data acquisition rate and communications performance. In this work, we develop a constrained multimodal sensing-aided communications framework where dynamic sensing and beamforming are performed under a sensing budget. Specifically, we formulate an optimization problem that maximizes the average received signal-to-noise ratio (SNR) of user equipment, subject to constraints on the average number of sensing actions and power budget. Using the Saleh-Valenzuela mmWave channel model, we construct the channel primarily based on position information obtained via multimodal sensing. Stricter sensing constraints reduce the availability of position data, leading to degraded channel estimation and thus lower performance. We apply Lyapunov optimization to solve the problem and derive a dynamic sensing and beamforming algorithm. Numerical evaluations on the DeepSense and Raymobtime datasets show that halving sensing times leads to only up to 7.7% loss in average SNR.

Dynamic Joint Communications and Sensing Precoding Design: A Lyapunov Approach

This letter proposes a dynamic joint communications and sensing (JCAS) framework to adaptively design dedicated sensing and communications precoders. We first formulate a stochastic control problem to maximize the long-term average signal-to-noise ratio for sensing, subject to a minimum average communications signal-to-interference-plus-noise ratio requirement and a power budget. Using Lyapunov optimization, specifically the drift-plus-penalty method, we cast the problem into a sequence of per-slot non-convex problems. To solve these problems, we develop a successive convex approximation method. Additionally, we derive a closed-form solution to the per-slot problems based on the notion of zero-forcing. Numerical evaluations demonstrate the efficacy of the proposed methods and highlight their superiority compared to a baseline method based on conventional design.

Sphere Precoding for Robust Near-Field Communications

Near-field communication with large antenna arrays promises significant beamforming and multiplexing gains. These communication links, however, are very sensitive to user mobility as any small change in the user position may suddenly drop the signal power. This leads to critical challenges for the robustness of these near-field communication systems. In this paper, we propose \textit{sphere precoding}, which is a robust precoding design to address user mobility in near-field communications. To gain insights into the spatial correlation of near-field channels, we extend the one-ring channel model to what we call one-sphere channel model and derive the channel covariance considering user mobility. Based on the one-sphere channel model, a robust precoding design problem is defined to optimize the minimum signal-to-interference-plus-noise ratio (SINR) satisfaction probability among mobile users. By utilizing the eigen structure of channel covariance, we further design a relaxed convex problem to approximate the solution of the original non-convex problem. The low-complexity solution effectively shapes a sphere that maintains the signal power for the target user and also nulls its interference within spheres around the other users. Simulation results highlight the efficacy of the proposed solution in achieving robust precoding yet high achievable rates in near-field communication systems.

Hybrid MIMO in the Upper Mid-Band: Architectures, Processing, and Energy Efficiency

As 6G networks evolve, the upper mid-band spectrum (7 GHz to 24 GHz), or frequency range 3 (FR3), is emerging as a promising balance between the coverage offered by sub-6 GHz bands and the high-capacity of millimeter wave (mmWave) frequencies. This paper explores the structure of FR3 hybrid MIMO systems and proposes two architectural classes: Frequency Integrated (FI) and Frequency Partitioned (FP). FI architectures enhance spectral efficiency by exploiting multiple sub-bands parallelism, while FP architectures dynamically allocate sub-band access according to specific application requirements. Additionally, two approaches, fully digital (FD) and hybrid analog-digital (HAD), are considered, comparing shared (SRF) versus dedicated RF (DRF) chain configurations. Herein signal processing solutions are investigated, particularly for an uplink multi-user scenario with power control optimization. Results demonstrate that SRF and DRF architectures achieve comparable spectral efficiency; however, SRF structures consume nearly half the power of DRF in the considered setup. While FD architectures provide higher spectral efficiency, they do so at the cost of increased power consumption compared to HAD. Additionally, FI architectures show slightly greater power consumption compared to FP; however, they provide a significant benefit in spectral efficiency (over 4 x), emphasizing an important trade-off in FR3 engineering.

Digital Twin Aided Channel Estimation: Zone-Specific Subspace Prediction and Calibration

Effective channel estimation in sparse and high-dimensional environments is essential for next-generation wireless systems, particularly in large-scale MIMO deployments. This paper introduces a novel framework that leverages digital twins (DTs) as priors to enable efficient zone-specific subspace-based channel estimation (CE). Subspace-based CE significantly reduces feedback overhead by focusing on the dominant channel components, exploiting sparsity in the angular domain while preserving estimation accuracy. While DT channels may exhibit inaccuracies, their coarse-grained subspaces provide a powerful starting point, reducing the search space and accelerating convergence. The framework employs a two-step clustering process on the Grassmann manifold, combined with reinforcement learning (RL), to iteratively calibrate subspaces and align them with real-world counterparts. Simulations show that digital twins not only enable near-optimal performance but also enhance the accuracy of subspace calibration through RL, highlighting their potential as a step towards learnable digital twins.

Digital Twin Assisted Beamforming Design for Integrated Sensing and Communication Systems

This paper explores a novel research direction where a digital twin is leveraged to assist the beamforming design for an integrated sensing and communication (ISAC) system. In this setup, a base station designs joint communication and sensing beamforming to serve the communication user and detect the sensing target concurrently. Utilizing the electromagnetic (EM) 3D model of the environment and ray tracing, the digital twin can provide various information, e.g., propagation path parameters and wireless channels, to aid communication and sensing systems. More specifically, our digital twin-based beamforming design first leverages the environment EM 3D model and ray tracing to (i) predict the directions of the line-of-sight (LoS) and non-line-of-sight (NLoS) sensing channel paths and (ii) identify the dominant one among these sensing channel paths. Then, to optimize the joint sensing and communication beam, we maximize the sensing signal-to-noise ratio (SNR) on the dominant sensing channel component while satisfying a minimum communication signal-to-interference-plus-noise ratio (SINR) requirement. Simulation results show that the proposed digital twin-assisted beamforming design achieves near-optimal target sensing SNR in both LoS and NLoS dominant areas, while ensuring the required SINR for the communication user. This highlights the potential of leveraging digital twins to assist ISAC systems.

A Dataset Similarity Evaluation Framework for Wireless Communications and Sensing

This paper introduces a task-specific, model-agnostic framework for evaluating dataset similarity, providing a means to assess and compare dataset realism and quality. Such a framework is crucial for augmenting real-world data, improving benchmarking, and making informed retraining decisions when adapting to new deployment settings, such as different sites or frequency bands. The proposed framework is employed to design metrics based on UMAP topology-preserving dimensionality reduction, leveraging Wasserstein and Euclidean distances on latent space KNN clusters. The designed metrics show correlations above 0.85 between dataset distances and model performances on a channel state information compression unsupervised machine learning task leveraging autoencoder architectures. The results show that the designed metrics outperform traditional methods.

Sensing-Aided 6G Drone Communications: Real-World Datasets and Demonstration

In the advent of next-generation wireless communication, millimeter-wave (mmWave) and terahertz (THz) technologies are pivotal for their high data rate capabilities. However, their reliance on large antenna arrays and narrow directive beams for ensuring adequate receive signal power introduces significant beam training overheads. This becomes particularly challenging in supporting highly-mobile applications such as drone communication, where the dynamic nature of drones demands frequent beam alignment to maintain connectivity. Addressing this critical bottleneck, our paper introduces a novel machine learning-based framework that leverages multi-modal sensory data, including visual and positional information, to expedite and refine mmWave/THz beam prediction. Unlike conventional approaches that solely depend on exhaustive beam training methods, our solution incorporates additional layers of contextual data to accurately predict beam directions, significantly mitigating the training overhead. Additionally, our framework is capable of predicting future beam alignments ahead of time. This feature enhances the system's responsiveness and reliability by addressing the challenges posed by the drones' mobility and the computational delays encountered in real-time processing. This capability for advanced beam tracking asserts a critical advancement in maintaining seamless connectivity for highly-mobile drones. We validate our approach through comprehensive evaluations on a unique, real-world mmWave drone communication dataset, which integrates concurrent camera visuals, practical GPS coordinates, and mmWave beam training data...