channel modeling

This paper addresses the lack of a unified stochastic geometry framework for performance evaluation of Integrated Sensing and Communication (ISAC) systems. We propose the first comprehensive stochastic geometry framework that jointly models three ISAC integration levels: sensing-aided communication, communication-aided sensing, and fully joint sensing-and-communication. By integrating spatial point processes—including Poisson point processes, Matérn hard-core processes, and Poisson cluster processes—alongside ISAC signal models and unified performance metrics (e.g., Cramér–Rao bound, achievable rate, detection probability), we rigorously characterize the spatial randomness of nodes and scatterers/occluders in terrestrial, aerial, and vehicular networks. A systematic review of over 100 studies reveals fundamental mechanisms by which spatial randomness governs the communication–sensing trade-off. We further identify key limitations of existing models in modeling dynamics, scalability, and cross-layer joint optimization, and outline concrete directions for future research.

Large-scale combinatorial optimization problems—such as resource allocation and trajectory design in 6G wireless networks—face critical challenges in dynamic heterogeneous environments, including poor real-time responsiveness, limited scalability, and insufficient user intent understanding. Method: This paper proposes a large language model (LLM)-driven framework for semantic understanding and structured reasoning. It integrates natural language modeling, solver-augmented co-reasoning, and solution verification, jointly leveraging deep reinforcement learning and semantic inference to enable end-to-end mapping from user intent to optimization models. Contribution/Results: Compared with conventional heuristic and deep reinforcement learning approaches, the framework achieves significant improvements in inference latency and cross-scenario generalization. It natively supports emerging paradigms such as low-altitude economy networking and intent-driven networking. Furthermore, the study systematically surveys LLM application architectures, representative use cases, open-source toolkits, and benchmark datasets in wireless networks—providing both theoretical foundations and practical guidelines for building trustworthy, scalable intelligent network optimization systems for next-generation wireless infrastructure.

Existing UAV channel models neglect the spatial correlation of ground user mobility trajectories, leading to inaccurate characterization of dynamic LOS/NLOS transitions in urban environments and thus distorted modeling of path loss and shadow fading. To address this, we propose a spatially consistent air-to-ground channel model that jointly leverages azimuth and elevation angles to partition LOS/NLOS probabilities—enabling coupled deterministic path loss and stochastic shadow fading modeling. Crucially, it reproduces spatially correlated LOS/NLOS transitions without requiring full 3D environmental reconstruction—a first in the literature. The method integrates geometry-driven simulation, probabilistic modeling, and spatially consistent parameterization. Evaluated across diverse urban scenarios, the model significantly improves fidelity in path loss and shadow fading representation, thereby enabling high-accuracy link outage probability analysis.

This study addresses the lack of accurate channel modeling for 3.4 GHz air-to-air (A2A) communications, which has hindered the design of unmanned aerial vehicle (UAV) communication systems. Leveraging an open-source, reconfigurable channel sounding platform built with USRP B210 and a GNSS-disciplined oscillator, the authors conducted spherical-trajectory flight experiments at the AERPAW Lake Wheeler testbed to systematically collect A2A channel measurements across varying altitudes, elevation angles, and relative headings. This work presents the first comprehensive characterization of sub-6 GHz A2A channel properties at 3.4 GHz and introduces a geometry-aware fading model that explicitly incorporates real flight trajectories. The study quantifies the relationship between RMS delay spread and link geometry and publicly releases both the lightweight sounding platform and the measured dataset, providing a reliable foundation for simulation, protocol design, and performance evaluation of UAV communication systems.

This work addresses unreliable coverage in UAV-assisted corridor communication networks caused by shadowing effects on air-to-ground links. To tackle this, we propose a one-dimensional finite point process modeling framework that jointly incorporates spatial randomness of UAV base stations (BSs) and empirically characterized shadow fading. Specifically, we unify the spatial distribution of UAV-BSs—modeled as either a binomial or homogeneous Poisson point process—with measured shadowing statistics. Under a maximum-received-power association policy, we derive a closed-form expression for the coverage probability and significantly reduce computational complexity via the dominant interferer approximation. The analytical model is validated against real-world air-to-ground channel measurements, achieving an average error of less than 8%. This study provides a verifiable, low-complexity theoretical foundation for reliable deployment and performance evaluation of UAV-BS networks.

This study addresses a critical gap in communication-aware robotic planning, where existing approaches commonly rely on channel-level metrics to predict end-to-end 5G throughput—a practice lacking empirical validation in private 5G deployments. Conducted in a shielded underground industrial environment, the work integrates commercial ray-tracing simulations, Gaussian process regression with a rational quadratic kernel, a mobile robotic platform, and off-the-shelf 5G user equipment to perform real-world measurements. It reveals for the first time that dynamic adaptation of MIMO spatial layers is the primary cause of systematic overestimation of throughput by conventional channel models, with ray tracing significantly overpredicting performance even in line-of-sight conditions. In contrast, a data-driven approach that directly learns end-to-end throughput reduces prediction error by approximately two-thirds and exhibits near-zero bias, demonstrating clear superiority over traditional channel-centric modeling.

Wireless channel modeling for communications and radar systems suffers from heavy reliance on high-quality labeled data, poor generalization, and limited physical interpretability. To address these challenges, this paper proposes a Sparse Bayesian Generative Modeling (SBGM) framework that explicitly incorporates physical priors. Specifically, it is the first to embed the inherent compressibility of wireless channels into a generative model, enabling online learning from compressed measurements. Physical constraints—derived from electromagnetic propagation principles—and parameterized channel representations are integrated to ensure model transparency and interpretability. Moreover, the method supports zero-shot transfer across antenna configurations and frequency bands without retraining. Experimental results demonstrate that SBGM achieves high-fidelity reconstruction of channel parameter distributions using only a small number of compressed measurements. This significantly reduces data acquisition and labeling overhead while markedly improving environmental adaptability and cross-scenario generalization performance.

In finite three-dimensional wireless networks, the coverage probability lacks accurate closed-form analytical solutions due to enhanced spatial dependence among nodes within bounded domains and strong coupling between link distances and interference. Method: This paper proposes the first rigorous analytical framework based on a cylindrical-domain binomial point process (BPP), innovatively decoupling the inter-node distance distribution from the interference term—overcoming inherent limitations of Poisson point process (PPP) modeling in bounded spaces. Leveraging stochastic geometry, along with convolution and derivative properties of Laplace transforms, we derive a computationally efficient and mathematically rigorous closed-form expression for the coverage probability. Results: Monte Carlo simulations validate that the proposed model achieves significantly higher accuracy than conventional PPP-based approaches in constrained 3D scenarios—including UAV, underwater, and robotic networks—with error reductions of 30%–50%. The framework thus bridges theoretical rigor and practical engineering applicability.

This study addresses the challenge of quantifying the value of task-oriented semantic information in wireless communications, where spectrum resources are limited and existing approaches struggle to account for complex spatiotemporal correlations. To this end, the paper proposes a Semantic Value of Information (SVoI) framework grounded in mutual information, which measures the reduction in uncertainty about unknown system states when leveraging historical semantic observations. This work establishes, for the first time, a unified metric that jointly incorporates semantic content, spatiotemporal dependencies, information timeliness, and channel conditions. Under a Gaussian Markov model, closed-form expressions for SVoI and its upper and lower bounds are derived, and the impacts of separable versus coupled spatiotemporal structures on semantic value are analyzed. Both theoretical analysis and simulations validate the efficacy of the proposed framework, offering an optimizable objective function for semantic-aware communication systems.

This work addresses the lack of high-fidelity, experimentally validated models for aerial–terrestrial cellular communication links, which hinders unmanned aerial vehicle (UAV) system design and mission planning. Leveraging the AERPAW Lake Wheel testbed, the study employs custom Android devices to collect enhanced key performance indicators (KPIs) over real 4G/5G networks, integrating spatial parameters such as flight altitude, slant range, elevation angle, and azimuth. By fusing free-space path loss theory with empirical measurements, the authors develop a joint physical- and application-layer empirical channel model. Lightweight machine learning techniques—including random forests, gradient boosting, and neural networks—are incorporated to predict spatial variations in KPIs. The resulting model accurately captures the characteristics of aerial–terrestrial links, offering a practical tool for simulation and system design of cellular-connected UAVs.

This work addresses the challenge of guaranteeing statistical end-to-end latency and accuracy quality-of-service in multi-cell edge intelligence systems under spatiotemporal uncertainties. To this end, the authors propose a joint wireless and computational resource pre-deployment optimization framework. By integrating Poisson point processes, queueing theory, and empirical AI inference workload measurements, they establish a unified stochastic modeling framework and, for the first time, derive an analytically tractable expression for end-to-end offloading latency. The resulting non-convex joint optimization problem is decomposed into convex subproblems, enabling the attainment of a globally optimal solution. The study further uncovers fundamental trade-offs among base station density, cell size, transmission latency, computational cost, and user fairness, and identifies a cost-efficient design regime in interference-limited scenarios.

The lack of high-quality, reproducible, publicly available over-the-air wireless datasets hinders research on 5G-Advanced/6G air-ground integrated networks. Method: Leveraging the AERPAW experimental platform, this work systematically constructs and open-sources the first multidimensional real-world dataset, acquired via software-defined radios mounted on unmanned aerial vehicles (UAVs). The dataset spans RF sensing, LoRaWAN, and 5G non-standalone (NSA) air-to-ground communications across diverse altitudes, flight trajectories, and environmental conditions. It integrates spectrum sensing, flying base station deployment, ray-tracing simulations, and empirical measurements for high-fidelity aerial channel characterization. Contribution/Results: The released dataset includes raw I/Q samples, propagation parameters, and standardized preprocessing scripts. It enables rigorous validation of propagation models, machine learning–driven air-interface optimization, and foundational research toward 6G space-air-ground integrated networks.