An adaptive hybrid indoor WiFi fingerprinting and propagation parameter estimation using RANSAC LASSO regression

This paper presents an indoor positioning method which based on wireless propagation model's estimation and hybrid weight k nearest neighbors (HWKNN). Firstly, the fingerprint data is optimized by RANSAC LASSO algorithm. Then, the optimized fingerprint data is used to estimate the position of each WiFi access point (AP) and the parameters of the wireless propagation model. At last, the signal strength of each reference point (RP) can be calculated by the wireless propagation model and the user's position can be estimated by using HWKNN method. Experimental results demonstrate that the proposed method provides a more accurate positioning result compared with traditional methods.