Propagation Map Reconstruction via Interpolation Assisted Matrix Completion

Constructing a propagation map from set of scattered measurements finds important applications in many areas, such as localization, spectrum monitoring and management. Classical interpolation-type methods have poor performance regions with very sparse measurements. Recent advance matrix completion has the potential to reconstruct measurements, but spatial resolution is limited. This paper proposes integrate interpolation exploit both correlation low rank structure map. The proposed method first enriches observations using interpolation, develops statistics error based on local polynomial regression model. Then, two uncertainty-aware algorithms are developed statistics. It numerically demonstrated that outperforms Kriging other state-of-the-art schemes, reduces mean squared (MSE) reconstruction by 10%–50% for medium large number