Architecture Design and Dynamic Resource Prediction for the Collaboration Between NFC and SAGIN

The collaboration between the near-field communications (NFC) and space-air-ground integrated network (SAGIN) is a potentially research direction for future development of wireless communications, providing deep integration of seamless wide-area coverage and high-precision near-field interaction. Particularly, benefited from the unique spherical waveform electromagnetic characteristics of NFC, the complementary advantages of such integration lead to the improvements in terms of the flexibility, reliability, and intelligence level of communication systems. In this article, we propose a layered collaboration architecture by means of analyzing the complementarity of related technologies, typical application scenarios, and key challenges. In addition, considering the high dynamism of communication nodes during network resource allocation, we study gated recurrent unit (GRU) and extreme gradient boosting (XGBoost) models to predict network resources in SAGIN-NFC networks preventing device access failures when accessing the network. Moreover, the method of dividing the dataset based on user activity levels is proposed to enhance prediction accuracy. The experimental comparisons demonstrate that the XGBoost model achieves lower prediction errors, validating the effectiveness of our proposed scheme.