Addressing the challenge of limited labels in Specific Emitter Identification (SEI), various Semi-Supervised Learning (SSL) approaches have been explored. However, these methods often neglect pervasive unknown signals in the real electromagnetic environment, thus restricting the model’s practical adaptability. To tackle this issue, we introduce Open-Set Semi-Supervised Learning (OSSL), a novel approach that excellently enhances the model’s robustness and generalization performance in complex electromagnetic environments.We present Fioma, an OSSL-SEI model distinguished by its innovative CenterOvA Loss. This loss function combines generative (Center) and discriminative (OvA) losses to optimize feature representations. Specifically, generative loss refines intraclass distributions through labeled samples, while the Exponential Moving Average (EMA) dynamically adjusts class radii for adaptability and stability. Discriminative loss considers inliers and outliers, achieving tight intraclass clustering and clear interclass separation. Through the innovative CenterOvA Loss, Fioma exhibits notable classification and clustering capabilities, even with limited labels.

Specific Emitter Identification, a critical technique for physical layer authentication in the Internet of Things, extracts Radio Frequency Fingerprints from signals affected by hardware impairment to discern between devices. The soaring development of the radio industry and the exorbitant expenses associated with signal labeling collectively necessitate the adoption of semi-supervised learning. This study proposes an approach employing curriculum triple consistency regularization to enhance Semi-Supervised Specific Emitter Identification. We incorporate constraints on weak-strong, branch, and semantic consistency, while implementing curriculum thresholds that dynamically filter pseudo-labels during training. Our approach achieves significant enhancements in classification and clustering performance, especially when faced with challenging training scenarios, while maintaining a lightweight codebase and low time and space complexity. Taking the 20-class ADS-B dataset with a 5% labeling ratio as an example, our approach achieved 77.80% accuracy and a Silhouette Coefficient of 0.272106, representing improvements of 13.00% and 0.138745, respectively.

With the increasing food demand for agricultural products and climate change pressures, the production of agricultural ecosystem performs more and more important to achieve the goal of sustainable development. The aim of this paper is to evaluate cropland production potential integrating cultivation patterns and types discrimination based on long-term crop dynamics using remote sensing approaches. First, crop cultivation patterns and types are identified using long-term vegetation dynamics information. Then the results are introduced into the models of cropland production potential estimation. Considering the influence of radiation, temperature, precipitation and soil conditions on crop growth for various crop types in different regions, the cropland potential production is estimated step by step. And finally show the results of four kinds of potential productivity: light, light-temperature, climatic and cropland potential productivity.

Soil chemical properties are the most important limiting factors in agricultural production. With the development of traditional agriculture towards modern precision agriculture, how to acquire soil information rapidly is a major issue for all the soil related workers. In this paper, hyperspectral technique is involved so as to discover the potential of using hyperspectrum in soil properties monitoring and estimation. First, soil properties investigation and sampling are conducted in the study area in the seedling growth stage of winter wheat. Then, soil samples were taken back to the laboratory to acquire soil chemical properties, such as soil organic matter, total N, available P and K, as well as their respective hyperspectral reflectance. The soil hyperspectral reflectance curves were filtered by wavelet transform in 6 scales using Matlab and obtain 6 low-frequency components and 6 high-frequency components, respectively. All these information are considered to estimate soil properties integrating multi-resolution analysis (MRA) and multivariate analysis (MVA). Results show that, the above methods improve the estimation accuracy obviously and could be extended to the estimation using hyperspectral images.

Mapping high spatial-temporal resolution evapotranspiration (ET) over large areas is important for water resources planning, precision irrigation and monitoring water use efficiency. However, both the traditional field measurement and aerodynamic estimation mainly focus on obtaining local ET. Remote sensing data often can be used to retrieve large area instantaneous ET at low spatial resolution over region or global scale. Therefore, using traditional measurements and high resolution image data to generate high spatial-temporal resolution ET is becoming an important research direction. In this paper, the complementary relationship model (CR) was employed together with meteorological data to estimate actual ET, and the results were validated by lysimeter observation. Furthermore, CR model was combined with high resolution image, IKONOS data, to estimate instantaneous field scale ET and they also were transferred into daily ET. The cumulative evapotranspiration (ET) of winter wheat during the reproductive phase from March to June of 2011 was 469.12 mm, essentially corresponding to the annual precipitation in the Beijing area. The most high accuracy of estimated ET by CR model is also on May(R2=0.863, RMSE=0.103 mm). The transferred daily ET by self-preservation of evaporative fraction(EF) method were consistent with lysimeter measurements for all four months(R2=0.937, RMSE=0.668 mm). It was proved in this study that CR model can be used to estimate precision field scale ET with meteorological data and high resolution remote sensing data together in a region with limited ground data availability.