Effect of Radio Map Interpolation on Performance of Rank-Based Fingerprinting Algorithm

The paper deals with the experimental evaluation of the impact of radio map interpolation on the accuracy of rank-based fingerprinting algorithm as well as nearest neighbour algorithms. The performance of algorithms was tested on a sparse radio map created using dynamic data collection. The created sparse radio map was processed by kriging, linear and inverse distance weighting interpolation algorithms in order to increase the density of reference points. The impact of different interpolation algorithms used to create radio maps on localization algorithms was evaluated. Based on the achieved results it can be concluded that interpolation does have some impact on algorithms that estimate the position of mobile devices from multiple reference points.