Jian Wang

High‐frequency (HF) radio signals rely on ionospheric reflection and refraction to enable long‐range and transcontinental transmission, making them essential for military command, emergency response, and strategic communications. The received field strength is a key parameter for quantifying propagation conditions and optimizing HF circuit design and application. To seek the optimal model and its improvement method, therefore, this study evaluates the prediction performance of three models (REC533, ICEPAC, and VOACAP) using the Consultative Committee on International Radio (CCIR) data set and the International Telecommunication Union (ITU) recommended standardized procedure. The assessment covers multiple dimensions, including time, distance, operating frequency, season, solar activity, and the midpoint position of the path. The results show that REC533 achieves the best overall performance, exhibiting a near‐zero bias of 0.28 dB(μV/m) and the lowest standard deviation of prediction errors, 9.81 dB(μV/m), together with consistently high stability across all propagation scenarios. In contrast, ICEPAC and VOACAP exhibit systematic underestimation of approximately 4 dB(μV/m), with substantially larger dispersion exceeding 15 dB(μV/m), particularly for long‐distance circuits, at high frequencies, and in low‐latitude regions. These findings clarify the applicability and limitations of the three prediction models, provide quantitative guidance for HF spectrum planning, circuit design, and emergency communication support, and offer directions for future model refinement.

Tropospheric evaporation ducts can trap radio waves within a refractive layer, which may guide signals above 1 GHz, enabling beyond-line-of-sight transmission. This makes duct-assisted propagation attractive for maritime communications. The marine environment is characterized by complex hydrometeorological variability and frequent extremes, particularly tropical cyclones, which can perturb duct properties and degrade link reliability. This study develops a multivariate cyclone-aware nonlinear regression framework (CNRF) to estimate contemporaneous evaporation duct strength (EDS) by integrating high-resolution dropsonde observations with tropical-cyclone descriptors from the International Best Track Archive for Climate Stewardship (IBTrACS). The framework uses CatBoost, natural-gradient boosting (NGBoost), and a multilayer perceptron (MLP) as base learners, with a random forest (RF) serving as the second-stage nonlinear fusion model. Rather than relying solely on bulk physical parameterization, the framework aims to represent the nonlinear influence of tropical cyclone-related environmental factors on duct strength. Evaluated over 1996–2024, the CNRF attains a test-set R² of 0.791 and a root mean square error (RMSE) of 5.350 M-unit, corresponding to a 23.5% improvement in RMSE over the Naval Postgraduate School (NPS) numerical model. For Hurricane Fiona (2022), the model achieves an RMSE of 6.260 M-unit, and the inclusion of tropical cyclone descriptors improves RMSE by approximately 17.0% relative to a model that excludes tropical cyclone information. The proposed framework facilitates quantitative assessment of extreme-weather-driven duct variability and supports robust design and operation of duct-enabled maritime communication systems.

The peak height of the F2 layer (hmF2) is a key parameter describing the vertical structure of the ionosphere. It is important for high-frequency radio communication planning and space-weather background assessment, particularly at high latitudes. To improve long-term hmF2 prediction, an empirical model-guided CatBoost model is developed. Predictions from the SHU and E-CHAIM empirical models are incorporated as prior predictors, while spatiotemporal periodicity, solar-activity, and geomagnetic-activity indices are jointly considered to represent the primary drivers of hmF2 variability. A two-stage feature selection procedure, combining stability-based selection and correlation-based redundancy pruning, is employed to identify informative and nonredundant features. The proposed model achieves consistently lower errors than both empirical models. Relative to SHU and E-CHAIM, the proposed model achieves relative root mean square error (RMSE) reductions of 22.67% and 17.70%, respectively, and relative mean relative error (MRE) reductions of 23.21% and 17.79%, respectively. Consistent improvements are observed across different time periods, seasons, and solar-activity conditions. The largest performance gains occur during spring and years of high solar activity. These results demonstrate that integrating empirical-model information with machine learning effectively improves the representation of hmF2 variability at high latitudes. The proposed model provides an effective empirical model-guided approach for long-term spatiotemporal prediction of hmF2 in high-latitude regions.

As a critical parameter characterizing the vertical structure of the ionosphere, the peak height of the F2 layer (hmF2) plays an essential role in frequency selection and signal propagation in high‐frequency (HF) communication systems. Accurate prediction of hmF2 is particularly important in high‐latitude regions. Based on observations from nine ionospheric stations located above 60°N, we analyze hmF2 variations and evaluate the prediction performance of four models: three IRI‐2020 sub‐models (IRI‐BSE, IRI‐AMTB, and IRI‐SHU) and the Empirical Canadian High Arctic Ionospheric Model (E‐CHAIM). The results show that: (a) hmF2 increases during high solar activity years and decreases during low solar activity years, exhibiting distinct diurnal and seasonal variations. (b) Among the four models, E‐CHAIM performs the best. Compared with the Bilitza–Sheikh–Eyfrig model (BSE), Altadill–Magdaleno–Torta–Blanch model (AMTB), and Shubin models (SHU), E‐CHAIM reduces the mean absolute error by 10.18, 11.85, and 1.77 km, and the root mean square error by 11.82, 12.61, and 1.98 km, respectively. (c) The performance of BSE and AMTB improves with increasing solar activity intensity, and the BSE model shows a more pronounced advantage during high solar activity years. It should be noted that the evaluation is conducted using monthly median hmF2 values, meaning the results primarily reflect the long‐term climatological performance of these models, while short‐term disturbances (e.g., geomagnetic activity) are largely suppressed. These findings deepen the understanding of long‐term variations in the high‐latitude ionosphere and lay a foundation for the long‐term planning of communication systems.

Accurate forecasting of the Maximum Usable Frequency (MUF) remains a major challenge in high-frequency (HF) wireless communications, where ionospheric variability directly impacts channel availability. This study develops a high-accuracy MUF forecasting framework using empirical data from 17 HF circuits across China to enhance the precision of frequency selection in HF communication. The main contributions are: (1) This study applies the Granger causality test for the first time to identify statistically significant causal relationships between MUF and space weather parameters. Lagged variables demonstrating causal influence are then selected as input features for the forecasting model. (2) In addition to traditional solar activity indices and geomagnetic data, this study also considers the influence of interplanetary parameters (Bz and solar wind) on MUF forecasting performance. (3) This study is the first to utilize the interpretable TabNet model for MUF forecasting, enabling quantification of input feature importance. Hyperparameter tuning is performed using simulated annealing to enhance model performance. Compared with the VOACAP and ITU-R REC533 models, the proposed TabNet model demonstrates high stability and forecasting accuracy across seasonal and hourly timescales. Overall, across these established models, it achieves a 35.34% improvement in performance. This work offers a fresh perspective on frequency optimization in HF communications, laying the groundwork for future advancements in adaptive communication strategies.

Broadcasting plays an important role in offshore information transmission because of its stability and long-distance propagation characteristics. To improve propagation prediction in the offshore environment of the South China Sea, this study proposes a path loss prediction model based on Support Vector Regression and verifies the performance with measurement data. Compared with ITU-R P.1546, the proposed model achieves an improvement in prediction accuracy of up to 55.50%. The findings of this research can help to give a clearer understanding of the path loss characteristics of the Frequency Modulation broadcasting signal in the offshore environment, which provides a more reliable foundation for radio wave transmission in offshore environments.

Developing accurate prediction method for propagation loss is significant for enabling efficient deployment and improved performance of wireless communication systems. To further enhance the accuracy of radio wave propagation prediction in the VHF and UHF bands, this paper proposes a prediction method of propagation loss based on the concept of data-model intelligent fusion. The proposed method is grounded in measured observational data and deeply integrates the domain knowledge of traditional propagation models, including ITU-R P.1546, ITU-R P.2001, and the Irregular Terrain Model (ITM), with a multilayer perceptron (MLP) optimized by the Seagull Optimization Algorithm (SOA) and a mean squared error (MSE) minimization fusion algorithm. By combining the interpretability of physical models with the strong nonlinear modeling and generalization capabilities of artificial intelligence, a high-accuracy and highly adaptive hybrid method is constructed, suitable for complex propagation environments. Based on measured data, a propagation loss prediction method was developed. Comparative analysis with existing models demonstrates that the proposed method achieves an average root mean square error (RMSE) of 5.12 dB, representing improvements in prediction accuracy of 82%, 44%, and 85% over the ITU-R P.1546, ITU-R P.2001, and ITM models, respectively. In addition to significantly enhancing prediction accuracy, the proposed method effectively reduces the environmental parameter measurements and the computational burden associated with building a high-accuracy propagation model, highlighting its strong engineering applicability and potential for practical deployment.

High‐frequency (HF) communication is an important means of strategic and emergency communication, playing an irreplaceable role in both military and civil applications. The maximum usable frequency (MUF) is a key parameter controlling HF signal propagation, and its real‐time and accurate forecast has long been a major challenge in engineering. To address this issue, this study proposes a transferable short‐term MUF forecasting model based on mutual information for feature selection and the Neural Basis Expansion Analysis with Exogenous Variables architecture. The model integrates intelligent approaches such as deep learning and adopts a phased, progressive hyperparameter optimization strategy (random search and Tree‐structured Parzen Estimator) to enhance model performance. Experiments conducted on the Changchun–Jingyang circuit demonstrate that the proposed model achieves an RMSE of 1.66 MHz on the test set, representing a 27% improvement over baseline models. Further transfer experiments on the Chongqing–Hainan, Xinxiang–Suzhou, and Guangzhou–Hainan circuits demonstrate that the proposed model achieves an improvement of over 25% compared with the baseline models, confirming its strong generalization and adaptability. The experimental data we used span four seasons—spring, summer, autumn, and winter—and cover both high and low solar activity years, ensuring the reliability of the model. Overall, the proposed MUF forecasting model achieves high accuracy while exhibiting strong transferability, providing an effective technical solution for real‐time HF communication applications under complex environmental conditions.