An efficient RF fingerprint positioning algorithm based on uneven grid layout

A novel meshing method based on uneven grid layout is proposed, which is designed for grid-based radio frequency (RF) fingerprint positioning and construct the uneven grid layout with a multistage hexagonal network model. We employ the uneven grid layout in RF fingerprint positioning framework which using the minimization of drive testing (MDT) to collect the RF fingerprint information and construct the fingerprints. Performance evaluation of the uneven grid layout is conducted in two LTE network scenarios, homogeneous and heterogeneous networks. The results show that our meshing method can reduce the size of the correlation database over the regular layout, and the position accuracy have at most 16% improvement over the conventional regular RF fingerprint layout.