Deep learning to robustify a geometric interpretation of trilateration for 3D RSS-based localization

Received Signal Strength (RSS) is ubiquitous in wireless communications. Despite the low accuracy, it is still attractive because of the simplicity and the ready availability in nearly every wireless system without any additional hardware or software required. This paper develops a geometric interpretation of trilateration in RSS-based 3D localization, which is presented in a previous paper but in 2D scenarios. In addition, to correct the final estimates, an iterative Maximum Likelihood (ML) estimator for position estimation is presented. The Artificial Neural Networks (ANNs) are then applied in estimating the related parameters to robustify the performance of the algorithm. When compared to earlier methods, simulation results demonstrate a considerable boost in performance.