Dimension Conversion Approach for Indoor Three Dimensional Radio Environment Map design

This study examines the effectiveness of employing the texture-patch transformation (TPT) approach compared to two-dimensional (2D) and three-dimensional (3D) approaches for generating indoor 3D radio environment maps (REMs). Two interpolation algorithms- K-nearest neighbor (K-NN) and inverse distance weight (IDW) are used for the evaluation. It’s crucial for the dynamic generation of REMs to be fast enough to keep up with the continuously received signal strength data from sensors in real-time, especially for dynamic spectrum access in television (TV) white spaces through cognitive radio networks. This research focuses on analyzing a symmetric vertical profile of an indoor 3D RSS dataset, which serves as the basis for implementing TPT in REM design. The findings indicate that the TPT method significantly reduces computation time (CT) compared to 2D and 3D approaches for IDW and K-NN. While TPT exhibits moderate accuracy in terms of root mean square error, relative recovery error, correlation coefficient, and best-fit line analysis, it’s premature to conclude that TPT lacks good accuracy based solely on interpolation results from a 30% training dataset. TPT demonstrates a favorable trade-off between accuracy and CT. The analysis reveals that 3D K-NN yields the highest accuracy among the six algorithms tested and covers a volume of 22,525 \(m^{3}\) $m^3$ of TV grey space indoors. The optimal values for IDW order, K-NN parameter, and all results are determined through cross-validation. TPT approached framework could be applicable in high-speed real-time 3D dataset reconstruction or indoor 3D REM generation using 2D interpolation approache, where the dataset follows symmetrical pattern.