A deep learning-based approach is presented that optimizes environmental input construction for accurate channel path loss prediction and validate that the input feature construction improves channel prediction accuracy and provides guidance for future intelligent channel prediction.
Abstract
Deep learning based wireless channel prediction can benefit from full exploitation of propagation features, yet how to construct effective inputs for accurate channel prediction remains under-explored. This paper presents a deep learning-based approach that optimizes environmental input construction for accurate channel path loss prediction. A campus measurement campaign is conducted to obtain datasets, where satellite image and semantic layers are aligned at a unified environmental granularity. Using a fixed network backbone and identical training protocol, results show that channel prediction accuracy is sensitive to spatial coverage. Overly local patches are insufficient, while enlarging coverage to include propagation-relevant surroundings yields error reduction with diminishing returns. It is further found that orientation normalization, implemented by aligning the Tx-Rx direction to a fixed reference axis, improves performance by reducing geometric variability. Building on this spatial extent, Shapley-based attribution and ablation analysis indicate that buildings and roads have dominating impacts, vegetation offers complementary information, and location-related descriptors mainly contribute through interactions. The results validate that the input feature construction improves channel prediction accuracy and provides guidance for future intelligent channel prediction.
Wireless localization is expected to play a key role in future communication systems by providing location-aware services and supporting efficient network operation. However, existing deep learning (DL)-based localization methods often suffer from limited generalization when the deployment environment changes, since they tend to learn environment-specific propagation patterns. To address this issue, this article proposes an environment-aware generalized wireless localization framework that jointly exploits wireless channel, base station (BS) geometric information, and environmental information. Irregular city structures are represented by voxel-based occupancy maps, enabling explicit modeling of environmental factors that affect radio propagation. A transformer-based architecture is developed to comprehensively process wireless channel, geometric information of network nodes, and environmental information, thereby capturing the interaction between channel observations and surrounding urban structures. In addition, the proposed framework estimates a confidence map instead of directly regressing user equipment (UE) coordinates, which improves robustness under ambiguous propagation conditions. To support training and evaluation, we also develop an urban environment generator and a ray tracing-based channel simulator that produce large-scale datasets with physically consistent alignment between channels and 3-D environments. This framework enables systematic evaluation and robust localization in previously unseen urban environments.
Y. Noh, Kae Won Choi· IEEE Internet of Things Jour...· 0 citations
Channel state information (CSI) acquisition, reconstruction, and prediction are fundamental yet costly tasks in modern MIMO-OFDM wireless systems. Direct coefficient-level prediction of raw CSI is fragile in realistic propagation environments, since small spatial perturbations, local scattering changes, and phase variations can cause large errors in the complex channel domain. However, the underlying wireless propagation field still contains stable and predictable structures that can be exploited across time, frequency, antenna, and carrier dimensions. Motivated by this observation, we propose a JEPA-based field-layer world model (FWM) that learns a shared latent propagation state from multi-resolution CSI observations across the considered carrier bands and predicts its task-relevant evolution in the latent domain. The proposed FWM maps multiple CSI observation resolutions to a shared latent propagation-field space through scale-specific tokenizer heads. A latent prediction backbone is then trained to infer masked or future field states, while an incremental multi-scale alignment strategy allows new observation scales to be incorporated without retraining the entire model from scratch. For downstream reconstruction, the predicted latent field is used as a structured prior and combined with sparse current pilots. Experiments on single-band and cross-band reconstruction demonstrate improved symbol detection and, more notably, substantial beamforming gains despite modest NMSE improvements, indicating that FWM captures task-relevant spatial propagation structure beyond coefficient-wise CSI fitting.
Yuzhi Yang, Brahim Mefgouda, Han Zou et al.· 0 citations
Path loss (PL) prediction is critical for wireless networks and most data traffic originates indoors. Effective network planning and base station deployment in high-frequency bands of 5G/6G systems necessitate addressing the accuracy–efficiency tradeoff. We propose a deep learning framework that combines dilated convolutions with spatial attention mechanisms to enhance PL prediction accuracy. Our model adaptively extracts multiscale environmental features while focusing on site-specific PL through attention-driven localization. Experimental results demonstrate superior performance over conventional models. In addition, the proposed architecture eliminates input size constraints and enables direct fine-tuning with measurement data, offering practical advantages for facilitating real-world deployment.
Chen-Yu Zhu, Stefanos Bakirtzis, Zi-Yi Xie et al.· IEEE Antennas and Wireless P...· 0 citations
This paper presents a multi-task spatiotemporal deep learning framework designed for zero-shot geographic generalization in predictive channel modeling for low Earth orbit (LEO) satellite communications at Q-band (39 GHz). Unlike many existing methods that rely on location-specific training, the coordinate-based long short-term memory (LSTM) architecture processes historical sequences of atmospheric and geographic data spanning 60 min to simultaneously predict weather conditions and excess path loss (EPL) for 5 h, incorporating inherent Gaussian uncertainty quantification. The proposed framework is designed to improve geographic generalization, which remains a challenge for existing machine learning approaches in satellite communications. A comprehensive evaluation conducted across various European climates demonstrates exceptional dual performance, achieving a root mean square error (RMSE) of 0.026 dB at the training locations and an average error of 0.213 dB across ten entirely unseen European cities, demonstrating strong generalization and representing a substantial improvement over existing methods. The framework maintains consistent prediction performance across all untrained European cities, while the uncertainty estimation mechanism provides calibrated confidence measures that may support operational decision-making. Furthermore, the coordinate-based inference strategy enables prediction at previously unseen locations without location-specific retraining, making the proposed framework a promising approach for large-scale predictive link adaptation in future LEO satellite networks. Although the proposed framework achieved promising results on ITU-R model-generated datasets, further validation using independent real-world satellite link measurements is needed to confirm its practical applicability under operational conditions.
Salma A. El-Sayed, Micheal Ibrahim, Hadia M. Elhennawy et al.· Journal on Wireless Communic...· 0 citations
Results show that the proposed method consistently outperforms classical path-loss modeling, interpolation, Kriging, and encoder–decoder baselines, especially when only a small fraction of measurement locations is available, support propagation-prior-guided model-aided learning as a practical approach for low-cost IoT radio-map construction and wireless signal management.
Ming-Kun Lu, A. Taparugssanagorn· IEEE Access· 0 citations
A channel knowledge map (CKM) provides location-specific channel priors and can reduce the overhead of real-time channel state information (CSI) acquisition for 6G environment-aware communications. In practice, CKM generation is often constrained by sparse and noisy measurements due to the high cost of wireless data collection. In this paper, we propose PDiff, a physics-informed conditional diffusion framework for CKM generation under sparse observations. Specifically, PDiff incorporates an analytical free-space propagation prior to capture the dominant distance-dependent attenuation trend, and combines it with environmental geometry, observation masks, and sparse observations as structured conditions. These conditional inputs guide the generation process with explicit propagation-aware, environmental, and measurement constraints. To improve inference efficiency, we further develop Prop-Cache, a training-free acceleration mechanism that reuses slowly varying intermediate features across denoising steps to reduce redundant computation during sampling. Experiments on RadioMapSeer demonstrate that PDiff outperforms a wide range of baseline methods for CKM generation.
Yu Chen, Jiao Chen, Jian Tang et al.· IEEE Transactions on Network...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
