| Title |
An Improved Super-resolution Reconstruction Method for RF Fingerprint Maps using Neural Networks |
| Authors |
정성훈(Sunghoon Jung) ; 임채훈(Chaehun Im) |
| DOI |
https://doi.org/10.5573/ieie.2023.60.10.3 |
| Keywords |
Indoor localization; RF fingerprint map; Super resolution; Sparse to dense; Neural network |
| Abstract |
This paper proposes an improved super-resolution reconstruction technique for RF fingerprint maps using neural networks. The proposed technique utilizes a neural network trained on the characteristics of the indoor wireless signal environment and fingerprint map to convert low-resolution fingerprint maps into high-resolution maps suitable for precise localization. Simulation results demonstrate that the proposed scheme enhances localization performance by efficiently reconstructing a high-resolution fingerprint map compared to previous studies. This approach can be utilized to meet the requirements of high-precision localization performance while reducing the effort required for fingerprint map data collection. |