Over the past few years, WIFI indoor localization has attracted a lot of attention, particularly due to people spending increasing amounts of time indoors. Nonetheless, achieving accurate indoor localization across multiple floors remains challenging due to signal fluctuations. Consequently, inconsistencies arise between the physical distance and the received signal strength indicator (RSSI) distance among different points. Moreover, in multi-floor environments, the received RSSI fingerprint generates a high-dimensional sparse matrix which results in high computational costs. This paper presents a novel multi-floor indoor localization method called Distance Metric Learning K-nearest neighbors Localization (DML-KNN) aiming to enhance floor identification rate and localization accuracy. To mitigate the effects of signal fluctuations, we propose a novel method for correcting the RSSI fingerprint with distance metric learning. This method improves the accuracy of the KNN algorithm, both in the floor classification phase and localization phase. The simulation and experimental results demonstrate that our proposed DML-KNN method achieves high localization accuracy.

LEO (low earth orbit) mega-constellation have the characteristics of large scale and low transmission delay. With the increase in the size of the constellation, the switching of the astrolabe has intensified, and the dynamic and management expenses of network topology have also increased sharply. In order to reduce the frequency of the switching of the ISL (Inter satellite link), a topology design algorithm based on the visibility of the star is proposed. After comparison analysis, the reduction of the switching frequency has greatly improved the stability of satellite network topology.

Significant Wave Height (SWH) is a critical parameter for ocean observation. SWH retrieval using GNSS-R data faces challenges including difficult feature selection, insufficient temporal dependency modeling, and limitations due to single-modality data. This paper proposes a novel Multimodal Transformer Network (MultTransNet) to enhance the accuracy of GNSS-R SWH retrieval. To optimize the feature set, we designed an XGBoost-based iterative feature selection module that effectively eliminates redundant features. To capture complex temporal dependencies and global context, the model employs a Transformer encoder utilizing its self-attention mechanism. Furthermore, to overcome the constraints of single-modality data, we innovatively fused 2D DDM image data with 1D auxiliary parameters, enabling multi-source information integration. Simulation results show that the Transformer architecture reduces Root Mean Square Error (RMSE) by 8.91% and increases Correlation Coefficient (CC) by 4.05% compared to a conventional Deep Neural Network (DNN) model. More significantly, the proposed multimodal algorithm further improves retrieval accuracy by 27.05% (RMSE reduction) and 7.21% (CC increase) compared to its single-modality Transformer counterpart, demonstrating superior performance, especially in complex sea-state conditions.