Limited Sampling Spatial Interpolation Evaluation for 3D Radio Environment Mapping

Area of Science:

• Wireless communication networks
• Spectrum sharing technologies
• Radio environment mapping

Background:

• Modern wireless networks require agile spectrum sharing due to increasing densification and diversification.
• Radio environment maps (REMs) are essential for spectrum utilization characterization and adaptive resource allocation.
• Accurate spatial interpolation methods are needed to estimate REMs effectively.

Purpose of the Study:

• To evaluate the performance of two established spatial interpolation algorithms for 3D radio environment data.
• To assess interpolation accuracy in real-world indoor and outdoor (UAV-based) scenarios.
• To analyze the impact of data sampling on 3D spectrum occupancy characterization.

Main Methods:

• Collected 3D spatial data indoors using a mechanical system and outdoors using an unmanned aerial vehicle (UAV).
• Applied two established interpolation algorithms to 2D planes at various altitudes and to limited samples (regions of interest).
• Analyzed algorithm performance using Kriging error standard deviation (STD) and the STD of distances between measurement and estimated points.

Main Results:

• Achieved a minimum error of -9.5 dB with a sampling ratio of 21%.
• Performance varied between indoor and UAV-based outdoor scenarios.
• Identified challenges in interpolation performance and spatial region of interest analysis.

Conclusions:

• The study provides insights into the performance of 3D data interpolation for REMs in diverse environments.
• Results highlight challenges and facilitate future development of 3D spectrum occupancy characterization.
• Findings are applicable to both indoor and UAV-based wireless network planning and optimization.