Radio Map Extrapolation Using Compensated Empirical CDF Under Interference-Limited Observations

Abstract

The radio map has attracted attention as a promising tool for accurately estimating radio propagation and determining communication parameters. Conventional radio maps are mainly used in wireless systems with a single target transmitter. However, in dense transmitter environments, such as cellular systems, interference-limited observations may become dominant, and thus, the received signal power from a target transmitter may be calculated to be higher than the actual value because of survivorship bias. As a result, the coverage area may be overestimated and communication parameters inappropriately determined. In this paper, we propose a radio map extrapolation method for multiple-transmitter environments. The proposed method compensates for the empirical cumulative distribution function (CDF) of the target power by taking into account the number of missing data in each mesh. The median target power was extrapolated based on the compensated empirical CDF. The emulation results using datasets over the 3.5 GHz band show that the proposed method can improve the mean error of the target power from 7 [dB] to 8 [dB], compared with the multiple imputation (MI) method, a non-compensated radio map, and kriging-based spatial interpolation.

Keywords

• Radio map
• missing data
• extrapolation
• empirical CDF